Skip to main content

Full text of "The ants of Polynesia (Hymenoptera: Formicidae)."

See other formats

Pacific Insects Monograph 14: 1-109 

15 August 1967 


(Hymenoptera : Formicidae) 

By Edward O. Wilson and Robert W. Taylor 1 


Abstract : All of the known Polynesian species are reviewed, including those newly dis- 
covered in several extensive, hitherto unstudied collections made during the past 30 years. 
Also, the ecological results of a field trip to Samoa in 1962 are presented. A total of 
83 species is recorded, of which 34 are " tramp " elements (introduced by modern human 
commerce from various parts of the tropics), 16 are endemics in western Polynesia, 27 are 
elements continuously distributed from Melanesia which may be native to part or all of 
their Polynesian range, and 6 are of uncertain zoogeographic status. The New Zealand 
fauna, containing 31 species, is reviewed separately in an Appendix. Nine endemic Poly- 
nesian species are described as new : Strumigenys mailei, Pheidole aana, P. atua, Vollenhovia 
pacifica, Rogeria exsulans, Adelomyrmex samoanus, Camponotus navigator, C. rotumanus, Poly- 
rhachis rotumana. All of these, except the last two, are from the Samoan Islands. Few if 
any ant species are native to the islands east of Rotuma, Samoa, Tonga, and New Zealand. 
This part of Polynesia has been populated by tramp species, which today present evidences 
of an approach to an equilibrial species density. The tramp species also show signs of 
some competitive replacement as well as of considerable population fluctuation. The avail- 
able information on the histories and ecology of individual species is considered with spe- 
cial reference to these phenomena. 

In the following account we wish to describe a fauna in the conventional taxonomic 
sense. It is also our aim to put down enough locality records to comprise a first chapter 
in what is certain to be a rapidly moving historical sequence. The ants of Polynesia are 
unusual in that nearly half of the species have been introduced into the Pacific by modern 
human commerce within the period of intrusion by European civilization, spanning no 
more than 400 years. East of Rotuma, Samoa, Tonga, and New Zealand there are ap- 
parently no naturally occurring ant species at all ; the introduced " tramp " species play 
an increasingly dominant role as one moves progressively to the Society Islands, Tuamotu 
Islands, Marquesas, and, finally, Hawaii. Such a newly assembled fauna can logically be 
expected to be in a state of flux. The situation is exceptionally interesting from the view- 
points of the ecologist and evolutionist, and for this reason we have undertaken to describe 
it as thoroughly as possible. Elsewhere (Wilson & Taylor 1967) we have described some 
of the more interesting theoretical implications of the distributions. The New Zealand 
fauna has recently been reviewed by Brown (1958) and later authors (see Taylor 1961), 

1. Present address of Taylor: Division of Entomology, C. S. I. R. O., Canberra, A. C. T., Australia. 

2 Pac. Ins. Mon. 1 ' 

and will not be covered in detail here. In the appendix we have provided a synopsis of 
the New Zealand species, adding some additional information that has come to light since 
Brown's paper. The New Zealand fauna, especially its native component, is very different 
from that of the rest of Polynesia and can conveniently be treated as a separate unit. 


The first collections of Polynesian ants were made in Samoa, Tonga, and the Ellice Is- 
lands in the 1860's and were described by Gustav Mayr (1866, 1870). They included several 
common Samoan endemics, such as Ectomomyrmex insulanus and Vollenhovia samoensis as 
well as many of the tramp and widespread Indo-Australian species that occur in western 
Polynesia today. Carlo Emery (1899, 1914a) added a few additional species from Samoa 
and Hawaii collected by Schauinsland and Silvestri. In 1899 Auguste Forel described the 
extensive collections made for Perkins' Fauna Hawaiiensis. In this first monograph of the 
ant faunula a of Polynesian archipelago, the largely homovectant character of the Hawaiian 
species was made evident. In 1928 Felix Santschi published a thorough study of the Samoan 
ants based on the collections made by P. A. Buxton and G. H. E. Hopkins during the 
London School of Hygiene and Tropical Medicine Expedition. In the same year L. Evelyn 
Cheesman published, in collaboration with W. C. Crawley, an account of her collections 
from the Society Islands and Marquesas. William M. Wheeler, the most prolific author 
on the Polynesian ants, began his studies in 1908 with a report of a small collection from 
the Society Islands. In 1932 he added more complete records from the Society Islands 
and described the large collections assembled by A. M. Adamson and E. P. Mumford in 
the Marquesas during 1929-30. Shortly thereafter Wheeler commenced work on the collec- 
tions of the B. P. Bishop Museum in Honolulu, producing a spate of descriptions and 
species lists until his death in 1937. Wheeler was fortunate to acquire the remarkably 
thorough collections gathered by E. C. Zimmerman during the Mangarevan Expedition of 
1934. Employing the sweep net, Zimmerman had sampled the faunas of many islands which 
had never before been investigated, from Rapa to Bora Bora and Henderson. Wheeler 
also reported on the extensive collections being accumulated in Hawaii by resident ento- 
mologists there. In the past four decades many entomologists, including F. X. Williams, 
R. H. Van Zwaluwenburg, O. H. Swezey, E. C. Zimmerman, E. H. Bryan, Jr., M. R. Smith, 
and others, have paid varying degrees of attention to the Hawaiian ants, especially those 
of economic importance, and have published a great deal of useful information, mostly 
in articles in the Proceedings of the Hawaiian Entomological Society. 

In 1940 O. H. Swezey and E. C. Zimmerman collected ants intensively on Samoa. T. E. 
Woodward added excellent Berlese funnel and hand collections in 1956, during a collect- 
Fig. 1. The major islands of the Western Samoan group, upper Savai'i, lower Upolu. The 

following localities are indicated on SAVAI'I : Alagaogao 2 ; Aopo 9 ; Asau 5 ; Faga 24 ; 

Falealupo 1; Gagaifoumauga 6; Ga'utaivai 16; Letui 11; Lotogo 3; Manase 13; Matautu 14; 

Matega 7; Mt. Matafa 20; Mt. Olomanu 21; Palauli 22; Papa 17; Patamea 18; Puapua 23; 

Safune 12 ; Salailua 8 ; Salelologa 26 ; Samalaeulu 19 ; Sili 15 ; Taga 10 ; Tuasivi 25 ; Vaisala 4. 

On UPOLU: Afiamalu 40; Aleisa 29; Apia 34; Fagaloa Bay 54; Faleapuna 51; Falevao 52; 

Fasitoouta 27 ; Lake Lanoto'o 33 ; Lauli'i 46 ; Le Mafa 53 ; Malololelei 39 ; Matautu 35 ; Moata'a 

42; Poutasi48; Sa'agaf ou 45 ; Salani 50; Saluafata 47; Satalo 49; Sliding Rock 32; Tanumalala 

28; Tapatapao 31; Ti'avea 55; Togitogiga 41; Utumapu 44; Vaea 37; Vaiala 36; Vailima 38; 

Vaipoto 30; Vivase 43. 

4 Pac. Ins. Mon. 14 

ing tour in association with R. A. Cumber, then resident entomologist at Upolu. This 
new, unstudied material provided the stimulus for our own work. We soon realized the 
crucial need to obtain a detailed picture of Samoa, since that archipelago contains the 
largest number of probable native species of any Polynesian archipelago and has undoubted- 
ly served as a principal source for insular faunulae to the east. Accordingly, in 1962 R. 
W. Taylor and his wife, Wendy, made a 5-week visit to Savai'i, Tutuila, and Upolu to obtain 
more thorough collections and gather ecological information. The success of these studies 
owes much to the assistance of G. Ettershank, then Entomologist to the Western Samoa 
Dept. of Agriculture. The Taylors visited many localities in all major habitats on Upolu 
and Tutuila, most of the time being spent working from a mountain cabin at Afiamalu in 
the center of Upolu. Taylor and Ettershank together covered Savai'i in an intensive 4- 
day program, during which almost all major roads were traversed with frequent collection 
stops. The interior of Savai'i could be entered at only 2 points, but lower elevation habitats 
were heavily sampled. Thanks to the cooperation of Dr G. Keys, then director of Apia 
Observatory, it was possible to operate a number of electrically powered Berlese funnels 
at Afiamalu, and additional funnels were run in Ettershank's laboratory. Following the 
Taylors' visit Ettershank continued intensive Berlese funnelling for several months, provid- 
ing us with many valuable records, especially from Savai'i and Tutuila. 

At the conclusion of the Samoan field studies we felt confident that a nearly complete 
faunal list could be made for the Samoan Islands. At the same time J. L. Gressitt turned 
over to us new collections accumulated in the Bishop Museum from other parts of Poly- 
nesia, including Rotuma, Tonga, the Danger Islands, Tokelau Islands, and Society Islands. 
Also, I. E. Efford made available the collection of ants made during the University of 
British Columbia Medical Expedition to Easter Island in 1965. 

In the present account an attempt has been made to record all collections of Polynesian 
ants, old and new, because of the possibility they provide of revealing short-term historical 
changes in the faunulae. For the same reason we have given either the date of capture 
or else a literature citation that will fix the approximate date of capture of most specimens 
collected on the islands. For economy, the following symbols relating to literature records 
are employed throughout : 

An asterisk before a locality means a literature record in one or more of the publications by 
W. M. Wheeler. Thus : 

* SOCIETY IS. means that all the ensuing records under the Society Is. are based on Wheeler's 
authority ; *Arne means that Arne in particular is based on Wheeler ; Papenoo Valley (1916 ; 
*1935) means that only the 1935 record is based on Wheeler. We have " spot-checked " many 
of Wheeler's determinations and become familiar with his species concepts. In case of dubious 
records we have either verified Wheeler's determinations or quoted them with qualification. 

A double asterisk before a locality means a literature record from Samoa by Santschi. Although 
we have not verified all of Santschi's determinations, we believe our knowledge of the Samoan 
fauna is sufficient to evaluate their reliability in most cases. 

A basic reference collection, including voucher series for many of our Samoan records 
as well as the holotypes of the new species, has been deposited in the Museum of Com- 
parative Zoology, Harvard University. A second comprehensive collection, including 

1967 Wilson & Taylor : Ants of Polynesia 5 

duplicate paratypes, will be placed in the B. P. Bishop Museum. A third collection, in- 
cluding paratypes and the bulk of the alcoholic material, and hence most of the voucher 
series of Taylor's records, will go to the Australian National Insect Collection at Entomology 
Division, C. S. I. R. O., Canberra, Australia. 


A classification of the 83 known Polynesian species according to origin is given in 
Table 1. The composition of the faunulae of the better known islands is given in Table 
2. In order to provide a crude measure of the degree of sampling, the number of persons 
who have made substantial collections on each of the islands is also given in Table 2. 
From these data, together with a consideration of distributions of individual species de- 
tailed later in the taxonomic part, the following generalizations have been drawn. 

(1) It is apparent that prior to the coming of man, few if any native species ranged 
east of New Zealand, Rotuma, Samoa, and Tonga. No certain endemics are known to 
occur in Polynesia east of these islands. Five species which prior to 1950 were considered 
to be endemic, namely Ponera swezeyi (Wheeler), Cerapachys silvestrii Wheeler, Epitritus 
wheeleri Donisthorpe, Monomorium rapaense Wheeler, and Plagiolepis mactavishi Wheeler, 
have been synonymized with species that occur elsewhere within or outside Polynesia. 
Four others, Amblyopone zwaluwenburgi (Williams) and Hypoponera zwaluwenburgi (Wheeler) 
of Hawaii, Smithistruma mumfordi (Wheeler) of the Marquesas Islands, and Oligomyrmex 
tahitiensis Wheeler of Tahiti, are still unknown elsewhere ; but their endemic status is cast 
into doubt by the fact that they belong to notoriously poorly collected and taxonomical- 
ly little known genera. Moreover, O. tahitiensis was described only from the sexual castes 
and cannot even be compared with most of the other Indo-Australian members of the 
genus, which are known only from the worker caste. Also, despite a plethora of subspecific 
and varietal names applied in the past literature to populations of species now living in 
the central and eastern Pacific, we have discovered only a single example of true geographic 
variation within these populations. The case is furthermore a relatively trivial one : the 
propodeal spines of workers and soldiers in samples of Pheidole umbonata Mayr from the 
Society Islands and Marquesas are slightly thicker, and the body color darker, than in 
similar samples from Samoa. 

Table 1. A Classification of the Polynesian Ant Species According to Origin. 
1. Endemic to one or more Polynesian archipelagoes : 

Ectomomyrmex insulanus, Ponera hi, P. woodward!, Strumigenys mallei, Pheidole aana, P. 
atua, Vollenhovia pacifica, V. samoensis, Rogeria exsulans, Adelomyrmex samoanus, Cam- 
ponotus navigator, C. rotumanus, C. flavolimbatus, C. conicus, C. nigrifrons, Polyrhachis 

Continuously distributed from Indo-Australian area into Polynesia : 

Prionopelta kraepelini, Platythyrea parallela, Ponera incerta, P. tenuis, Hypoponera confinis, 
H. punctatissima, Cryptopone testacea, Odontomachus simillimus, Anochetus graeffei, Eurho- 
palothrix procera, Smithistruma dubia, Strumigenys szalayi, Pheidole fervens, P. oceanica, P. 
umbonata, P. sexspinosa, Solenopsis papuana, Oligomyrmex atomus, Monomorium talpa, Tet- 
ramorium pacificum, T. tonganum, Rogeria sublevinodis, Iridomyrmex anceps, Tapinoma 
minutum, Technomyrmex albipes, Paratrechina minutula, Camponotus chloroticus. 

6 Pac. Ins. Mon. 

3. " Tramp species " : certainly distributed by recent human commerce : 

Hypoponera opaciceps, Trachymesopus stigma, Leptogenys maxilloso, Syscia silvestrii, Tri- 
choscapa membranifera, Strumigenys godeffroyi, S. lewisi, S. rogeri, Quadristruma emmae, 
Pheidole megacephala, Solenopsis geminata, Monomorium destructor, M. latinode, M. floricola, 
M.fossulatum, M.minutum, M.pharaonis, Triglyphothrix striatidens, Tetramorium caespitum, 
T. guineense, T. simillimum, Cardiocondyla emeryi, C. nuda, C. wroughtoni, Iridomyrmex 
humilis, Tapinoma melanocephalum, Anoplolepis longipes, Plagiolepis alluaudi, P. exigua, 
Paratrechina bourbonica, P. vaga, P. longicornis, Brachymyrmex obscurior, Camponotus varie- 

4. "Tramp species" intercepted in quarantine at Honolulu but not yet established in Polynesia : 

Brachyponera solitaria, Tetramorium caespitum, Wasmannia auropunctata. 

5. Uncertain status : 

Amblyopone zwaluwenburgi, Ponera swezeyi, Hypoponera zwaluwenburgi, Smithistruma mum- 
fordi, Oligomyrmex tahitiensis, Chelaner antarcticum. 

Table 2. 

Partition of the Faunulae of Individual Polynesian Islands According to Origin. 

Archipelago or 

Samoa (1938-62) 

1. Savai'i 

2. Upolu 

3. Tutuila 
Tokelau Islands 

4. Fakaofo (1924) 
Danger Islands 

5. Motu Kotawa 
Society Islands (1906-34) 

6. Bora Bora (1925-34) 

7. Mehetia 

8. Huahine 

9. Moorea (1906-34) 

10. Raiatea (1925-34) 

11. Tahiti (1925-34) 
Austral Islands 

12. Rimatara 

13. Raivavae 

14. Maria I. (N-E Islet) 

15. Rapa 

Gambier Islands (1934) 

16. Mangareva 
Marquesas (1920-29) 

17. Eiao 

18. Hatutu 

19. Ua Pu 

20. Ua Huka 

21. Hiva Oa 

Area in sq. No. of No. of 
miles collectors genera 





13 2 



No. of species 

Wilson & Taylor : Ants of Polynesia 

Archipelago or 

22. Tahuata 

23. Nuku Hiva 

24. Fatu Hiva 

25. Mohotane 

26. Pitcairn Island (1934) 

27. Henderson Island (1934) 

28. Flint Island (1934) 
Hawaii (1920-62) 

29. Hawaii 

30. Maui 

31. Molokai 

32. Oahu 

33. Kauai 

34. Nihoa 

35. French Frigate Shoals 

36. Laysan 

37. Wake 

38. Midway 

39. Kure (Ocean) 
Solomons (1916) 

40. Florida 

41. Malapaina 

42. Ugi 

43. Malaita 

44. San Cristoval 

45. Santa Ysable 





About 20 Indo-Australian ant species range to some point east of Rotuma, Samoa, and 
Tonga. Yet these too might easily have been transported into outer Polynesia by man. 
Iridomyrmex anceps (Roger), for example, one of the most widespread of the dolichoderines, 
was unknown until recent years from east of the Solomon Islands. In 1955 it was col- 
lected on Aitutaki, in the Cook Islands, and in 1956 at Nandi, the international airport 
community of Fiji. Intensive collecting has not yet revealed its presence in the intermedi- 
ately situated Samoan islands. The case for its establishment on Aitutaki through human 
commerce is therefore strong. Several other of the Indo-Australian elements in the central 
and eastern Pacific are certainly known to be tramp species, having been established in 
the New World as well. Three others Odontomachus simillimus Fr. Smith, Tetramorium 
pacificum Mayr, and Pheidole fervens Fr. Smith have been intercepted at quarantine stations 
in Hawaii (see Appendix I), and the last 2 have been taken in quarantine in New Zealand 
(R. W. Taylor, unpublished mss.). 

(2) The native species of western Polynesia are drawn almost exclusively from the 
Indo-Australian area. Almost all of the endemics have close relatives in Australia or 
Melanesia, mostly the latter. The single exception is Rogeria exsulans Wilson & Taylor, 
a small secretive ant belonging to a group which is otherwise known only from the Neo- 
tropical Region. 

8 Pac. Ins. Mon. 14 

(3) No one island contains all of the 34 tramp species known to occur in Polynesia, 
and most contain less than one-fourth of them. Several lines of evidence suggest that 
the species densities have stabilized. Competitive replacement has probably played some 
role in the stabilization. At least it has apparently limited the number of species of Phei- 
dole, Cardiocondyla, and Paratrechina occurring in individual islands. Distributions of 2 
or 3 species within these genera tend to be complementary. The clearest example involves 
the large, aggressive species of Pheidole. P. fervens, a widespread Indo-Australian species, 
is unknown from Samoa at the present time but is a dominant ant in the Society Islands. 
P. megacephala, a pantropical species of African origin, well known for its competitive in- 
teractions with other ant species, has the reverse distribution : it is dominant on Upolu 
(Samoa) but is rare or absent in the Society Islands. P. oceanica, another Indo-Australian 
element, replaces megacephala on Savai'i (Samoa) and occurs on Upolu only on the western 
side facing Savai'i ; it is furthermore relatively uncommon in the Society Islands. Else- 
where in Polynesia the complementarity among the 3 species is maintained. Fervens occurs 
on Tonga and Pitcairn ; it is occasional in the Marquesas and unknown from Hawaii. 
Megacephala is absent from Tonga and Pitcairn but is dominant on the Marquesas and 


Details of external anatomy of an ant worker are shown in Figure 2. The following 
terms, which are used in the taxonomic part, either are not found in the ordinary ento- 
mological glossary or else require special definition with reference to our study. 
Basal face (of propodeum). Dorsal surface of propodeum. 
CI. Cephalic index. (HWXlOO)/HL. 

Declivitous face (of propodeum). Posterior face of propodeum, extending to petiolar junc- 

EHB. Collected by E. H. Bryan, Jr. 

ECZ. Collected by E. C. Zimmerman. 

GE. Collected by G. Ettershank. 

HL. Head length. As conventionally used in myrmecology, the maximum distance, taken 
along the midline between the points of intersection of the midline with the lines 
drawn perpendicular to the midline, that touch the anteriormost and posteriormost 
parts of the head exclusive of the mandibles (the posteriormost and anteriormost 
points need not be on the midline). 

HW. Head width. When the head is viewed full face, the maximum width taken per- 
pendicular to the midline. This measurement is made across the eyes in the male 
but above or below the eyes (if these break the profile) in the queen and worker. 

Mesosoma. The alitrunk, or middle tagma, including the thorax and propodeum together. 

Microreticulum. A very fine reticulum in the cuticular sculpturing, where the reticular di- 
ameters are on the order of 0.01 mm. This is one particular form of the microsculp- 
ture generically referred to as " shagreening " by many authors. 

NK. Collected by N. L. H. Krauss. 

OHS. Collected by O. H. Swezey. 

Propodeal junction. Line of junction of the basal and declivitous faces (q. v.) . 

RWT. Collected by R. W. Taylor. 

Wilson & Taylor : Ants of Polynesia 




-. o: o_ <C 

00 CL O fc 

Fig. 2. Rhytidoponera metallica (Fr. Smith), worker from Sydney, Australia ; labelled 
to demonstrate the major terminology used in this study. 

SI. Scape index=(SLxlQO)/HW. 

SL. Scape length. The maximum length of this appendage that can be taken. 

Subopaque. Four arbitrary degrees in the spectrum of " shininess " are recognized : shin- 
ing, feebly shining, subopaque, opaque. The first and last are self-explanatory, and 
the middle 2 are meant to be (subjectively) 2 equidistant intermediate steps. Thus 
subopaque means mostly opaque but with a few points of reflection under ordinary 
reflected light. 

TEW. Collected by T. E. Woodward. 

Acknowledgments: We are indebted to our colleague William L. Brown, Jr., who with 
characteristic generosity supplied much vital information from his notebooks on the Pone- 
rinae and Dacetini. We are also very grateful to George Ettershank and Thomas E. Wood- 
ward, who turned over their Samoan collections, together with valuable ecological notes ; 
to J. L. Gressitt, who made possible the loan of the Bishop Museum collections ; to 
George Hunt, who in preparing for a field trip to Polynesia tested our keys and descriptions ; 

10 Pac. Ins. Mon. 14 

to Jonathan Reiskind for information on the taxonomy of Cardiocondyla ; to Mr and 
Mrs W. Lidgard and the officials of the Apia Observatory for assistance rendered to Taylor 
during his visit to Samoa ; and to I. E. Efford for the gift of his collection of Easter 
Island ants. Dr and Mrs Ettershank provided generous aid in many ways during the 
Samoan trip. The research has been supported by Grant No. GB 1634 from the U. S. 
National Science Foundation. The Samoan field studies were part of an extensive 5-month 
study tour which also included Fiji, Australia and New Guinea. This was carried out by 
Taylor under the auspices of the Committee on Evolutionary Biology, Harvard University ; 
the Bache Fund Committee, American Academy of Arts and Sciences ; and the Society of 
Sigma Xi-RESA. 


1. Gaster attached to mesosoma (alitrunk) by a "waist" consisting of 2, more or less 

reduced and nodiform segments ? 

Gaster attached to mesosoma by a " waist " consisting of a single segment, which 
may be nodiform, erect or inclined scale-like, or even prostrate and more or less 
hidden by overhanging gaster 3 

2. Pygidium flattened at apex, the flattened part bordered with denticles arranged in 

rows on each side ; underside of head with a strong carina on each side running 
forward from posterior corner ; frontal carinae closely approximated and vertical, 
so that the antennal insertions are exposed when head is examined in facial view 

Cerapachyinae (Syscia silvestrii) 

Pygidium simple ; no distinct carinae on underside of head extending forward from 
posterior corners ; frontal carinae usually horizontal, obscuring antennal insertions 
in facial view Myrmicinae 

3. Sting well developed and functional, usually extended and visible in dead specimens 


Sting absent, or vestigial and not extensible 4 

4. Seventh sternite rolled into a short ventro-apical cone with a round orifice (with 

or without a coronula of tiny hairs) that serves as a nozzle for a defensive acid 
spray ; not to be confused with cloacal orifice, which is more dorsal and normal- 
ly hidden Formicinae 

Gastric apex lacking such a spray-ejecting cone ; defensive secretion a viscous fluid, 
ejected through a slit-like orifice Dolichoderinae 


Head of bizarre form illustrated in figs. 19 and 20, with mandibles linear and 
inserted near its midline 7 

Head differently shaped, with mandibles inserted at its anterior corners 3 

Large sized (HW about 2mm) dark reddish-brown species with petiolar apex 
drawn into an acute conical spine (fig. 19) Odontomachus simiUimus 

Smaller (HW about 1 mm) golden brown species with petiolar summit a nar- 
row transverse ridge (fig. 20) Anochetus graeffei 

Petiole depressed, articulated over its whole posterior surface with postpetiole 
(fig. 4) 4 

Wilson & Taylor : Ants of Polynesia 

Articulation between petiole and postpetiole narrow, petiole usually with a 
distinct transverse posterior face (e.g., fig. 6) 5 

Mandibles short, closing tightly against clypeus, their apical borders distinct 
and occupied by 3 large teeth, of which the middle is shortest ; basal border 
of mandible edentate (fig. 4) Prionopelta kraepelini 

Mandibles linear, strongly projecting beyond clypeus when closed, their inner 
borders armed with a number of bipartite teeth (fig. 3)..Amblyopone zwaluwenburgi 

Mandibles falcate, very slender and strongly curved, lacking distinct teeth ; 
when closed there is an extensive gap (with an area much greater than that 
of mandibles themselves) between their inner borders and anterior clypeal 
border (fig. 18) ; pretarsal claws pectinate Leptogenys maxillosa 

Mandibles differently shaped, usually triangular, and with distinct teeth ; when 
closed there is little if any gap between them and clypeus; pretarsal claws 
simple, or with a single median tooth < 

Head, mesosoma and node very roughly punctate-rugose, dorsal aspects of 
postpetiolar and 1st gastric tergites densely and finely arched-striate ; entire 
body with strong greenish or purplish metallic reflections ; inferior margins 
of pronotum each armed with a strong acute tooth Rhytidoponera metallica 

Sculpturation unlike that described above, dorsal aspects of postpetiole and 
1st gastric segments never striate ; color ranging from black to pale yellow- 
ish brown, without metallic reflections; inferior pronotal margins rounded 

Petiolar node distinctly longer than broad in dorsal view ; body almost entire- 
ly lacking erect hairs none break its dorsal outline except at gastric apex ; 
pretarsal claws each with a distinct median tooth Platythyrea parallela 

Petiolar node usually distinctly broader than long in dorsal view, occasionally 
almost as long as broad ; body with abundant erect or suberect hairs break- 
ing its dorsal outline; pretarsal claws simple, lacking a median tooth 8 

Declivitous face of propodeum and posterior face of node heavily striate, 
striae usually transverse, though sometimes partly longitudinal on upper 
parts of node ; mesepisternum divided by a transverse suture into anepister- 
nal and katepisternal plates Ectoraomyrmex insulanus 

Declivitous face of propodeum and posterior face of node smooth and shin- 
ing, at most with a few transverse striae on their lower parts ; mesepisternum 
entire, not divided by a horizontal suture 9 

Mandible elongate triangular, the angle between its posterior and masticatory 
borders obtuse, approximating 120. Masticatory border with 5 or 6 distinct 
strong teeth (figs. 7, 8) 10 

Mandible broadly triangular, the angle between its posterior and masticatory 
borders approximating 90. Masticatory border with more numerous small 
teeth or minute denticles (e. g., fig. 9) 11 

Small (HW<0.5 mm) pale brown species, entirely lacking compound eyes ; an- 
tenna bearing a distinctly 4-segmented club Cryptopone testacea 

Larger (HW> 1 mm) dark brown species, with small but distinct compound 
eyes and lacking a distinctly segmented antennal club Trachymesopus stigma 

Clypeus with a small but distinct anteriorly directed conical median tooth 
(figs. 10, 13) p 

Pac. Ins. Mon. 

Clypeus lacking an erect median tooth, its mesal surface often carinate, but 
carina never extended apically as a tooth 13 

Scapes long, when laid back along head they cleat ly surpass median occipital 
border by a distance equal to 1-1. 5 X their maximum thickness ; scape index 
98-100 ; mesometanotal suture not represented on mesosomal dorsum as a 
sharply incised line, though it may be represented as a shallow obtuse con- 
cavity which does not break the underlying puncturation Ponera woodward! 

Scapes shorter, when laid back along head they barely attain median occipital 
border, or fail to do so by up to 0.5 X their maximum thickness ; scape index 
84-91 ; mesometanotal suture clearly incised on mesosomal dorsum as a 
sharp line, which interrupts the underlying puncturation Ponera loi 

Compound eyes completely lacking ; subpetiolar process without a fenestra (fig. 
17); antennal club indistinctly 5-segmented Hypoponera zwaluwenburgi 

Compound eyes present, though often represented by a single minute facet ; 
subpetiolar process with or without an elliptical anterior fenestra (figs. 9, 
14) ; antennal club usually segmentally undifferentiated or distinctly 4-seg- 
mented ; in cases where it is indistinctly 5-segmented the subpetiolar process 
is f enestrate 14 

Subpetiolar process with an elliptical anterior fenestra and a pair of bilateral 
posteroventral denticles (figs. 9, 13) ; compound eyes consisting of 1 or rare- 
ly 2 facets; maxillary palpi 2-segmented ; larvae when present ' ''' " 
of "doorknob-shaped" tubercles on abdominal dorsum 15 

Subpetiolar process a simple lobe-like structure without a fenestra or postero- 
lateral denticles (figs. 15, 17) ; compound eyes usually with more than 3 
facets ; maxillary palpi 1 -segmented, larvae with 2 pairs of dorsal abdominal 
" doorknob " tubercles 17 

Mesometanotal suture lacking on mesosomal dorsum ; antennal club distinctly 
4-segmented ; pupae naked, lacking cocoons Ponera tenuis 

Mesometanotal suture distinctly incised on mesosomal dorsum ; antennal club 
indistinctly 5-segmented; pupae when present enclosed in yellowish cocoons. ..16 

Larger medium brown species (HW 0.30-0.45 mm) with relatively broad head, 
its maximum width 0.78-0.84 X its median length Ponera incerta 

Smaller pale yellowish-brown species (HW 0.32-0.33 mm) with relatively nar- 
row head, its maximum width 0.76-0.78X its median length Ponera swezeyi 

Antennal scapes short ; when laid back along head they fail to attain median 

occipital border by a distance greater than their maximum diameter 

Hypoponera punctatissima 

Antennal scapes longer, clearly attaining or surpassing median occipital border... 18 

Dorsum of pronotum strongly shining and lacking puncturation ; mesepisternum 
almost completely outlined by a fine raised carina (fig. 14) ; petiolar node 
in side view distinctly narrowed dorsally Hypoponera confinis 

Dorsum of pronotum dully shining, with a close cover of very fine punc- 
tures ; mesepisternum not outlined by a carina ; anterior and posterior faces 
of petiolar node in side view almost parallel Hypoponera opaciceps 

Wilson & Taylor : Ants of Polynesia 


Antennae with 7 segments or less ; head cordate in front view and often 
bearing 2 or more conspicuous scale-like hairs ; postpetiole commonly (but 
not always) with bunches of whitish, spongiform material 2 

Antennae with at least 8 segments ; except in large-headed soldier caste of 
Pheidole, in which the head is somewhat cordate, the head in front view 
is always subrectangular to elliptical ; scale-like hairs not present on head 
and spongiform appendage never developed on postpetiole 11 

Mandibles short, thick and serially dentate ; when fully closed, engaging along 
their entire masticatory margins and leaving no appreciable interspace 3 

Mandibles linear, elongate, with only 3 or 4 spiniform teeth at or near their 
apices ; when closed, only these teeth engage, and a broad open space is 
framed between mandibular shafts (Dacetini pan.) 6 

Eyes large, dorsolaterally placed, and conspicuous when head viewed directly 
from front; large, HW>1.2mm; dark brown Eurhopalothrix procera 

Eyes very small, laterally placed, and not visible when head viewed directly 
from front; HW<0.8mm; light to medium reddish brown (Dacetini part.) ...4 

Mandibles short-triangular, with distinct, transverse basal borders ; hairs of 
dorsum of head proper limited to a single erect short clavate pair on vertex 
Trichoscapa membranif era 

Mandibles long-triangular, without transverse basal borders ; vertex and occiput 
with more than 1 pair of specialized erect spatulate or clavate hairs (8 in 
undamaged specimens), along with a ground pilosity of subreclinate spatulate 
hairs (genus Smithistruma) 5 

Smaller species, HL<0.60 mm ; diagonal length of mesosoma in side view<0.65 
mm Smithistruma dubia 

Larger species, HL>0.60mm; diagonal length of mesosoma in side view 
>0.65 mm Smithistruma munif ordi 

Antennal funiculus with only 3 distinct segments (small species with strongly 
bowed mandibles; head covered with large-orbicular squamiform hairs) ... 
Quadristruma emmae 

Antennal funiculus with 5 distinct segments, of which the 2nd and 3rd are 
small (genus Strumigenys) / 

Mandible with 2 preapical teeth (i. e., teeth near apex other than apical fork) 
Strumigenys rogeri 

Mandible with a single preapical tooth 8 

Sides of head abruptly indented at large convex eyes, which are oriented more 
or less anteriorly ; antennal scrobes obsolescent, scapes curving to fit sides 
of head in repose ; larger form, HW>0.53 mm (Samoa, originating in Mela- 
nesia) Strumigenys szalayi 

Sides of head not abruptly indented at eyes, and eyes oriented primarily to 
sides ; antennal scrobes well developed and receiving folded antennae in 
repose; smaller forms, HW<0.53 mm 9 

Mesosoma, especially pronotum, with very abundant long subreclinate or arch- 
ing fine hairs, giving dorsum a woolly appearance ; also a few long flagellate 

Pac. Ins. Mon. 

hairs Strumigenys godeffroyi 

Mesosoma with only a few long, erect hairs and sparse, short subreclinate 
under-pilosity 10 

Rear face of propodeum bearing spongiform lamellae ; 1st gastric segment 
bearing, just posterior to postpetiolar spongiform appendage, a dense fringe 
of long, reclinate, posteriorly directed hairs (known from Hawaii only) ... 
Strumigenys lewisi 

Lacking spongiform lamellae and fringe of hairs described above (Samoa 
only) Strumigenys mailei 

Antenna 9-segmented, including a robust 2-segmented club longer than entire 
remainder of funiculus ; propodeum bluntly dentate 12 

Antenna 10-, 11-, 1 2-segmented ; if terminal club is very large and 2-jointed, 
then propodeum is unarmed 13 

A very small species ; HW of minor worker only about 0.28 mm, that of sol- 
dier only about 0.46 mm (Samoa) Oligomyrmex atomus 

A larger species; HW of unknown worker castes perhaps 2x as great as in 
atomus (known only from sexual castes from Tahiti) ... Oligomyrmex tahitiensis 

Antenna 10-segmented, with a distinct 2-segmented club 14 

Antenna 11- or 12-segmented, with a distinct 1- or 3-segmented club, or else 
terminal joints not forming a distinct club 15 

Workers markedly polymorphic, the smallest with HW not less than 0.50 mm ; 
eyes containing more than 30 ommatidia Solenopsis geminata 

Workers monomorphic, with HW not exceeding 0.35 mm ; eyes containing less 
than 10 ommatidia Solenopsis papuana 

Anterior clypeal border bearing 4 conspicuous teeth ; a large recurved accessory 
tooth present near base of mandible and well behind masticatory border ; 
antennal club 1 -segmented ; small, robust, heavily sculptured, dark brown 
(Samoa) Adelomyrmex samoanus 

Anterior clypeal border with at most 2 teeth ; accessory, basal tooth lacking 
on mandible ; antennal club either 3-segmented or not distinct from re- 
mainder of funiculus 16 

Petiole subtended by a smooth, very flat, ventrally rounded, translucent flange 
about as broad as depth of anterior peduncle of petiole ; slender, heavily 
sculptured species (Samoa only) 17 

Petiole at most subtended by a small thin knob placed at anteriormost part 
of ventral surface of anterior peduncle 18 

Body blackish brown; propodeum unarmed Vollenhovia paciflca 

Body brownish yellow ; propodeum bearing small but distinct right-angular 
teeth Vollenhovia samoensis 

Frontal lobes fused with median 1/3 or 1/4 of clypeus, which forms a dis- 
tinct shelf raised sharply from lateral portions of clypeus 19 

Frontal lobes clearly demarcated from clypeus in front ; center of clypeus not 
conspicuously raised as a separate element from remainder of sclerite 23 

Body very densely covered with long silvery hairs, many of which are bifid 
or trifid Triglyphothrix striatidens 

Body pilosity sparse to moderately abundant; hairs never divided 

Wilson & Taylor : Ants of Polynesia 

. HW<0.65mm; body nearly uniformly light reddish brown 

HW>0.70 mm ; abdomen (gaster) very dark brown ; remainder of body dark 

brown or light reddish brown 

. Body hairs sparse, thick and blunt ; posteroventral propodeal flange right-angu- 
lar ; anterior peduncle of petiole about as thick as long...Tetramorium simillimum 

Body hairs abundant, slender and acute ; posterior propodeal flange drawn 
into an acute tooth ; anterior peduncle of petiole much longer than thick 

Tetramorium tonganum 

. Body nearly uniformly blackish brown ; length of propodeal spine much 
greater than thickness of anterior peduncle of petiole ... Tetramorium pacificum 

Body, except for abdomen (gaster) light reddish brown ; length of propo- 
deal spine roughly same as thickness of anterior peduncle of petiole 

Tetramorium guineense 

. Propodeum either smoothly rounded or, at most, armed with a pair of blunt 
processes forming angles of not less than 90 ; clypeus bicarinate (Monomo- 
rium) 2' 

Propodeum armed with a pair of acute teeth or spines ; clypeus not bicari- 
nate 31 

Eyes small, no more than 0.03 mm in greatest diameter and containing only 
2-4 ommatidia 

Eyes moderate in size, at least 0.06 mm in greatest diameter and containing 
more than 10 ommatidia 26 

Mesopleuron shagreened and opaque; paired propodeal angles about 100 

Monomorium f ossulatum 

Mesopleuron smooth and shining ; propodeal angles about 140.. .Monomorium talpa 

Antennal club not distinctly 3-segmented ; robust species with HW>0.65 mm 
(Rapa I. only) Chelaner antarcticus 

Antennal club distinctly 3-segmented; HW<0.60 mm ; 

Body entirely shagreened and opaque Monomorium pharaonis 

Body mostly or entirely smooth and shining 28 

Polymorphic, HW of smallest worker > 0.40 mm ; terminal antennal segment 
about as long as next 2 together 29 

Monomorphic, HW of largest worker<0.35 mm ; terminal antennal segment 
distinctly longer than next 2 together 30 

Seen from directly above, postpetiolar node is 1.4x broader than long; in 
side view postpetiolar node appears conspicuously larger than petiolar node 
Monomorium latinode 

Seen from directly above, postpetiolar node is l.lx-1.2x broader than long; 
in side view postpetiolar node appears approximately equal in size to pe- 
tiolar node Monomorium destructor 

Mesosoma (alitrunk) and pedicel clear yellow or at most light brownish yel- 
low, contrasting with dark reddish brown head and abdomen (gaster); 
seen from side, entire petiole, including peduncle, about 1.5x longer than 
deep Monomorium floricola 

Body rather evenly dark reddish brown to blackish brown ; seen from side 
entire petiole only about as long as deep Monomorium minutum 

Pac. Ins. Mon. 

Mesonotum and pronotum armed with pairs of well developed spines 

Pheidole sexspinosa 

Mesonotum and pronotum unarmed 32 

Seen from directly above, postpetiole nearly 2x as broad as petiole; dorsum 
of body devoid of standing hairs ; small (HW about 0.4 mm), slender, mo- 
nomorphic species 33 

Seen from directly above, postpetiole is at most 1.3x broader than petiole; 
dorsum of body bearing numerous standing hairs 35 

Mesometanotal groove absent or very weak Cardiocondyla nuda 

Mesometanotal groove distinct, especially in side view 3' 

Profile of mesonotum sloping gradually back towards mesometanotal groove; 
propodeal spines short and stout Cardiocondyla emeryi 

Mesonotal profile declining abruptly at mesometanotal groove ; propodeal 
spines moderately long and prominent Cardiocondyla wroughtoni 

Monomorphic ; pronotum coarsely rugoreticulate ; scapes in repose not reach- 
ing occipital border 36 

Dimorphic ; pronotum of small-headed minor worker either smooth and shin- 
ing or finely "shagreened"; scapes of minor worker in repose exceeding oc- 
cipital border 37 

Propodeal spiracle as wide as spine is long, or wider ; subpetiolar process ab- 
sent Rogeria sublevinodis 

Propodeal spiracle only 1/2 as wide as spine is long, or less; flange-like sub- 
petiolar process present Rogeria exsulans 

HW of minor worker about 0.4 mm, that of soldier about 0.8 mm ; both castes 
concolorous light reddish yellow Pheidole umbonata 

HW of minor worker 0.5 mm or greater, that of soldier 1.1 mm or greater; 
both castes light to dark reddish brown 38 

Occiput of soldier smooth and shining. Minor worker with mesosoma either 
entirely shagreened and opaque, or mesonotum nearly flat in side view, 
forming a smooth posterior continuation of curve of pronotum 39 

Occiput of soldier rugoreticulate and opaque. Minor worker with at least 
parts of pronotum, mesopleura, and sides of propodeum smooth and at least 
feebly shining; also, mesonotum strongly convex in side view 40 

When head of soldier is viewed in full face, rugae are seen to cover approx- 
imately anterior 3/4 of surface exclusive of mandibles ; pronotum of mi- 
nor worker entirely smooth and shining (distributed throughout Polynesia) 
Pheidole megacephala 

When head of soldier is viewed in full face, rugae are seen to cover only 
about 1/2 of surface exclusive of mandibles ; pronotum of minor worker 
covered with transverse rugae (Samoa only) Pheidole aana 

In soldier, when head is viewed from side the frontal region (between ends 
of frontal carinae and occiput) is seen to be distinctly concave ; pronotum 
of minor worker covered with transverse rugae (Samoa only) Pheidole atua 

Frontal region of soldier in side view convex ; most of pronotal surface com- 
pletely smooth (widespread in Polynesia) 41 

In soldier, area between antennal insertion and eye longitudinally rugose ; pro- 

Wilson & Taylor : Ants of Polynesia 

podeal spine of minor worker distinctly longer than greatest width of pro- 

podeal spiracle Pheidole oceanica 

In soldier, area between antennal insertion and eye rugoreticulate ; propodeal 
spine of minor worker only about as long as greatest width of propodeal 
spiracle Pheidole fervens 


Petiole armed with 2 large, laterally directed horn-like spines ; monomorphic, 

medium-sized, black (Rotuma only) Polyrhachis rotumana 

Petiole either unarmed or else bearing a single median tooth-like protuberance ... 2 

Antenna 9-segmented ; small, robust, brown Brachymyrmex obscurior 

Antenna 11- or 12-segmented 3 

Juncture of dorsal and basal faces of propodeum of workers (or minor workers 

if workers are polymorphic) drawn into an acute tooth-like protuberance ; 

medium-sized species 4 

Juncture of dorsal and basal faces of propodeum of minor workers rounded or 

at most obtusely angulate 5 

Petiolar node armed with a posterior median tooth-like protuberance (Tonga 

only) Camponotus conicus 

Petiolar node unarmed (Samoa only) Camponotus flavolimbatus 

Polymorphic ; medium-sized to large, HW (exclusive of compound eyes) of 

smallest worker > 0.80 mm 6 

Monomorphic; small to medium-sized, HW of largest worker not>0.75 mm in 

Head and mesosoma clear yellow , 

Head and mesosoma medium to dark reddish brown or darker 8 

Abdominal (gastric) tergites brownish yellow, their posterior margins slightly 

and gradually infuscated (Samoa, Tonga, Rotuma, Danger, Ellice) 

Camponotus chloroticus 

Posterior margins of abdominal tergites marked by subtriangular, medium brown 

infuscations sharply marked off from anterior parts, which are clear yellow 

(Hawaii only) Camponotus variegatus 

Basal face of propodeum strongly concave ; posterior margins of abdominal 

(gastric) tergites whitish, contrasting strongly with golden-brown anterior 

portions (Rotuma only) Camponotus rotumanus 

Basal face of propodeum straight or convex ; abdominal tergites concolorous 

blackish brown (Samoa and Tonga) 9 

Mesosoma clear reddish yellow ; dorsal face of propodeum flat in side view ; 

only 2 or 3 standing hairs breaking entire mesosomal profile (Tonga) 

Camponotus nigrif rons 

Mesosoma medium to dark reddish brown ; dorsal face of propodeum convex ; 

over 10 standing hairs breaking mesosomal profile (Samoa) 

Camponotus navigator 

Antenna 1 1 -segmented ; minute, robust, yellow formicine ants with almost com- 
pletely hairless mesosoma and large abdomen (gaster) distinctly longer than 

mesosoma in undistended state (Plagiolepis) 11 

Pac. Ins. Mon. 

Antenna 12-segmented ; not combining all of other characters cited above 1: 

, Scape exceeds occipital corner by less than length of 1st funicular segment 

(Hawaii only) Plagiolepis exigua 

Scape exceeds occipital corner by at least length of 1st 2 funicular segments 
combined (widespread and common in eastern and southern Polynesia as 

well as in Hawaii) Plagiolepis alluaudi 

Minute, HW approximately 0.40 mm, with mesosoma (alitrunk) completely 

devoid of standing hairs 13 

Either much larger (HW>0.60 mm), or else mesosoma bears numerous stand- 
ing hairs 14 

Head blackish brown, contrasting sharply with light brownish yellow anten- 
nae and gaster ; petiolar node rudimentary ; antennal scape surpassing occi- 
pital corner by distinctly more than length of the 1st funicular segment... 

Tapinoma melanocephalum 

Body concolorous light brown ; petiolar node inconspicuous but well devel- 
oped ; antennal scape surpassing occipital corner by distinctly less than 

length of 1st funicular segment Tapinoma minutum 

Body extremely thin and elongate; antennal scape at least 1.5X as long as 

head including closed mandibles 15 

Body of "average" to somewhat robust proportions ; antennal scape not more 

than 1.5X as long as head including closed mandibles 16 

Dorsum of mesosoma almost completely devoid of standing pilosity ; color 

yellow; mesonotum viewed from side weakly concave Anoplolepis longipes 

Dorsum of mesosoma bearing numerous long, erect hairs ; color grayish brown 
with occasional weak purplish reflections ; mesonotum viewed from side 

weakly convex Paratrechina longicornis 

Mesosoma devoid of standing pilosity (Hawaii only) Iridomyrmex humilis 

Mesosoma bearing at least several prominent standing hairs 17 

Anterior clypeal border emarginate 1 

Anterior clypeal border whole and convex 19 

Petiolar node well developed; mesosoma feebly shining Iridomyrmex anceps 

Petiolar node rudimentary; mesosoma densely shagreened and opaque 

Technomyrmex albipes 

HW not>0.40 mm Paratrechina minutula 

HW>0.50rnm ; - 

HW of great majority of workers in most nest series 0.45-0.62 mm; body 
usually light to medium reddish brown ; in case the workers are interme- 
diate in the preceding 2 characters to bourbonica (below) the 3-3- can be 

distinguished by the outer margins of the parameres being entire 

Paratrechina vaga 

HW of great majority of workers in most nest series 0.65-0.72 mm ; body 

usually dark brown; outer margins of $ parameres excised 

Paratrechina bourbonica 

Wilson & Taylor : Ants of Polynesia 

Subfamily PONERINAE 
Amblyopone zwaluwenburgi (Williams) Fig. 3. 

Stigmatomma (Fulakora) zwaluwenburgi Williams, 1946, Proc. Hawn. Ent. Soc. 12: 639, work- 
er (Type locality: Honolulu, Oahu). 
Amblyopone zwaluwenburgi: Brown, 1960, Bull. Mus. Comp. Zool. Harv., 122 (4): 213. 

DISTRIBUTION : HAWAII : Oahu, known only from 
the type material. f \ "N 

In his recent revision of the world Amblyoponini, Brown / ; 1 

(1960) states: "A. zwaluwenburgi is a minute species (un- ! 1 

der 2 mm total length) that was found in the soil of a \ P / 

sugar cane field at the Hawaiian Sugar Planters Associa- / \ ( >^ 

tion Experiment Station, Honolulu. I think it is likely that (. ' 

the species has been introduced into Hawaii from Mela- , 
nesia or the East Indies." We concur in this opinion. \. 

Like the smaller species of the genus Ponera, the celata ^ 

group of Amblyopone, to which zwaluwenburgi belongs, con- 
sists of tiny cryptobiotic species which are especially dim- Fig. 3. Amblyopone zwalu- 
cult to collect. As in the case of the Hawaiian "endemic" wenburgi (Williams), type 
Ponera swezeyi, recently recorded from Samoa, future col- ^^^2) 
lecting will probably unearth Amblyopone zwaluwenburgi out- 
side Hawaii, either in Samoa or in the chief faunal source areas west of Polynesia. 

Prionopelta kraepelini Forel Fig. 4. 

Prionopelta kraepelini Forel, 1905, Mitt. Mus. Hamburg 22 : 3, worker, queen. 

DISTRIBUTION: SAMOA: Upolu : Apia, winged greens, 31.111.1962 (RWT ace. 2305); 
4.IV.1962 (GE ace. 27). Vaivase, winged queens, 8.III.1962; 27.111.1962 (W. Lidgard). 

These 4 series, collected at light, compare well with queens in the MCZ collection from 
the Philippines. They are somewhat larger (HW 0.45 vs. 0.40 mm in the single measur- 
able Philippine specimen) but are still easily distinguishable from the much larger (HW 
ca. 0.65 mm) species majuscula Emery of New Guinea. P. kraepelini is very close to P. 
opaca Emery of New Guinea and Micronesia ; indeed the two may be synonymous (Brown 

Fig. 4. Prionopelta kraepelini Forel, worker from Micronesia. 

Pac. Ins. Mon. 

1960: 220). Kraepelini is distinguished by its weakly sculptured and partly shining body 
surface ; opaca is densely, coarsely punctate and almost completely opaque. 

The genus Prionopelta was previously unknown from the Pacific E of New Guinea and 
Micronesia. Moreover, P. kraepelini is evidently abundant on certain islands in Micro- 
nesia, but has never been collected on New Guinea. Its distribution is thus peripheral to 
that of opaca. These considerations, plus the fact that P. kraepelini is known only from 
a limited area in the environs of Apia, lead us to conjecture that it has been introduced 
into Samoa in recent times by human commerce. 

Platythyrea parallela (Fr. Smith) 

Fig. 5. 

Ponera parallela Fr. Smith, 1859, J. Linn. Soc., Zool. 3: 143, worker (Type locality: Aru). 
Platythyrea parallela : Wilson, 1958, Bull. Mus. Comp. Zool. Harv. 118 (3): 151, synonymy, 

Platythyrea pusilla var. pacifica Santschi, 1928, Insects of Samoa 5 (1) : 43, worker. Type 

locality : Apia, 500 m, Upolu, Samoa. New Synonymy. 

DISTRIBUTION: **SAMOA : Upolu : Apia (pacifica type series, collected in 1927). 
SOCIETY IS.: Tahiti: Patutua, 1925; Tautira, 1925 (Cheesman & Crawley 1928). 

Fig. 5. Platythyrea parallela (Fr. Smith), worker from lower Busu River, 
New Guinea. 

Platythyrea parallela ranges more or less continuously from tropical Asia to Samoa, ap- 
parently by-passing Fiji and New Caledonia. The species must be quite rare in Polyne- 
sia. Only 3 collections have been made, despite the fact that the workers are conspicu- 
ous and partially arboreal in their foraging habits, rendering them especially easy to 
capture in sweep nets. All of the records given above were made during 1925-1927. It 
is probably significant that no additional specimens were taken by Swezey and Zimmer- 
man or Woodward, during their collection tours in 1938-40 and 1956 respectively, or by 
the Taylors and Ettershank during 1962. It is reasonable to conjecture that the species 
is now extinct in the Samoan islands. 

Rhytidoponera metallica (Fr. Smith) Fig. 2. 

See Brown (1958) for current synonymy. 

1967 Wilson & Taylor : Ants of Polynesia 21 

DISTRIBUTION : Throughout Australia except far north ; New Zealand, introduced 
(Taylor 1961); Samoa (Mayr 1876). 

Mayr based his Samoan record on specimens in the collection of the Godeffroy Muse- 
um of Hamburg. If the labelling was correct in the first place, the Samoan metallica pop- 
ulation was undoubtedly introduced from Australia, and was probably ephemeral in na- 
ture ; no additional collections have been made in the intervening 90 years. 

Ectomomyrmex insulanus (Mayr) 

Fig. 6. 

Ponera insulana Mayr, 1876, J. Mus. Godeffroy, Hamburg 12: 87, worker (Type locality: 

DISTRIBUTION: SAMOA: Savai'i: Aopo, 170m, rain forest, berlesate, leaf mold and 
soil (RWT ace. 597). Gagaifoumauga, rain forest, berlesate, moss and bark from rotting 
log (GE ace. 55). Lotogo, 350 m, rain forest, strays under stone, in rotting log, and 
under epiphyte 35 m from ground on Ficus tree (RWT aces. 455, 459, 473). Mt. Olomanu, 
rain forest, berlesates, moss and soil from rotting logs (GE aces. 58, 59). Samalaeulu, 
coastal rain forest, berlesate, leaf mold and soil (RWT ace. 596). Upolu: Afiamalu, 700 
m, VI.1940 (ECZ) including winged queen, 8.VI.1940 ; disturbed rain forest, leaf mold 
and ground moss (RWT ace. 2324, GE aces. 46, 48) ; disturbed rain forest and adjacent 
Taro patch, claustral queens under stones 9-12.III. (RWT aces. 219, 266, 282), colonies 
from soil under stones (RWT aces. 245, 273, 317, 327) or rotting logs (RWT aces 220, 
316, 599). Malololelei Road, 650 m, ft, 18.VI.1940 (ECZ). Tiavi, 700 m (ECZ) 
This species is known only from Samoa where it was first collected in 1870. All re- 
cords are from Savai'i and Upolu. The 

A, ^s^U^%/,, ;/ ,, , , , ^ existence of an apparently endemic Ec- 

'<. tomomyrmex on Samoa is a remarkable 

^J/ N^~ fact in itself, since no other member of 
T^^/Jl^ t ^ le S enus rar >8 es E of the Solomon Is. 
v ' ^^ ^ E. insulanus is most closely approached 
^ within the genus by simillimus (Donis- 
thorpe) of New Guinea and aequalis Mann 


^ of the Solomons, which species are evi- 

jfe, dently cognate derivatives of a previously 

^X^WwlPi widespread Melanesian stock. E. insulanus 

r~V*~f\ \&f differs from both of the above species as 

follows. Size : insulanus is larger ; the 
Fig. 6. Ectomomyrmex insulanus (Mayr), worker entire series cited above ranges in HW 
from Afiamalu, Upolu. 1.32-1.39 mm, as opposed to 1.17-1.27 

mm in 2 series of aequalis and 1.14mm in 1 series of simillimus measured. Sculpturing: 
costulation and striation is everywhere more weakly developed ; insulanus workers can be 
recognized immediately by the fact that the 2nd gastric tergite is weakly punctate, where- 
as in all related species the tergite is strongly punctate-striate with its surface opaque. 

E. insulanus shows considerable variation within single nest series in the orientation of 
the striation on the posterior face of the node. At one extreme the striae are entirely 
transverse, while at the other extreme the upper parts of the sculptured area bear predom- 

Pac. Ins. Mon. 14 

inantly longitudinal striae. 

The preferred nesting site at Afiamalu appeared to be in the soil under logs or (more 
frequently) stones. Nests collected during March by the Taylors contained many pupae 
and a few large larvae ; occasional callow workers were also observed. 

Trachymesopus stigma (Fabricius) 

Fig. 7. 

Euponera (Trachymesopus) stigma var. quadridentata. : Emery, 1911, Gen. Insect. 118: 85. 

Wheeler, 1935, Occ. Pap. Bishop Mus. 11 (11) : 12. 
Euponera (Trachyponera) stigma v. quadridentata : Santschi, 1928, Insects of Samoa 5 (1) 


Fig. 7. Trachymesopus stigma (Fabricius), worker from Malololelei Road, 

Trachymesopus stigma : Wilson, 1958, Bull. Mus. Comp. Zool., Harv. 119 (4) : 355, syno- 
nymy, distribution; 1959, Evolution 13 (1): 128, fig. 2, distribution, ecology. 

DISTRIBUTION: SAMOA: Savai'i **Fagamolo (1923). **Safune (1924). Upolu : 
Apia, 300m (H. Swale). Malololelei Road, 500-600 m, $$, 18.VII and 8.VIII.1940 (OHS, 
ECZ). Tutuila: **Afone Trail (1923). **Center of island (1923). Fagatoga, 300m, &&, 
beating shrubbery, 25-28.VIII.1940 (OHS, ECZ). **Pago Pago (1923, 1924). Manua : Ta'u: 
winged queen, 20.11.1926 (A. F. Judd). 

This is one of the most widespread of all ponerine ants. In the New World it ranges 
from southern Florida through the Antilles and Central America to southern Brazil. In 
the Old World it is found continuously from southern China to the Caroline Is., northern 
Queensland, and throughout Melanesia (except New Caledonia) to Samoa. Although 
widely distributed on Samoa, it is evidently scarce there, being known chiefly from occa- 
sional collections of winged queens and males. T. stigma may indeed be extinct on Sa- 
moa since it has not been collected since 1940, in spite of the efforts of Woodward, the 
Taylors, and Ettershank. 

1967 Wilson & Taylor : Ants of Polynesia : 

Cryptopone testacea (Emery) Fig. 8. 

Cryptopone testacea : Wilson, 1958, Bull. Mus. Comp. Zool. Harv. 119 (4) : 360, distribu- 
tion, ecology. 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, partly cleared rain forest, strays 
under moss on logs (RWT aces. 222, 345) ; berlesate of ground moss, alate $, 9.IV.1962 
(GE ace. 53). 

Fig. 8. Cryptopone testacea (Emery), worker from Afiamalu, Upolu. 

The relationship and distribution of this small cryptobiotic ponerine have been discuss- 
ed at length by Wilson (1958). It is by far the most widely dispersed Cryptopone spe- 
cies, ranging continuously, and with no marked geographical variation, from Ceylon 
through Indonesia and Melanesia as far east as the Solomon Is. It has also been collected 
on Ponape, in Micronesia, but so far is unknown from Fiji or the New Hebrides. This 
species is probably not native to Samoa. It was found in the company of predominantly 
introduced species and was never encountered in undisturbed forest. 

Ponera incerta (Wheeler) Fig. 9. 

Pseudocryptopone incerta Wheeler, 1933, Amer. Mus. Novit. 672: 18, fig. 7, worker, 
(Type locality : Depok, Java) . 

Fig. 9. Ponera incerta (Wheeler), worker from Poentjak, Java. (After Tay- 
lor, 1967.) 

"4 Pac. Ins. Mon. 1 

Ponera incerta : Wilson, 1957, Bull. Mus. Comp. Zool., Harv. 116 (6) : 384. 
Ponera ratardorum Wilson, 1957, Bull. Mus. Comp. Zool., Harv. 116 (6) : 369, worker (Type 
locality: St. Pauls, Baining Mts., New Britain). 

DISTRIBUTION: Detailed synonymy and collection records of this species are given 
by Taylor (1967). P. incerta is known from Java, Yap I. (Carolines), New Britain, Es- 
piritu Santo (New Hebrides), and Santa Cruz and Guadalcanal (Solomons). In Polynesi 
it is apparently restricted to Samoa, where it has been collected in abundance in rain 
forest on Savai'i, Upolu and Tutuila. Taylor (he. cit.~) reports an apparent case of com- 
petitive displacement between this species and the rather similar P. tenuis, observed during 
1962 in disturbed rain forest at Afiamalu (Upolu). The 2 species have apparently parti- 
tioned the environment such that incerta was abundant nesting under moss on the ground 
or logs, while tenuis was significantly more abundant than incerta in moss on tree trunks, 
and almost wholly limited to the latter arboreal habitat. This apparent displacement was 
not observed at other localities, and may have been precipitated by the fact that both 
species were maintaining extremely dense populations at the Afiamalu site. Certain other 
native and exotic ant species were unusually abundant at the same place. 

Ponera loi Taylor Fig. 10. 

Ponera loi Taylor, 1967, Pac. Ins. Mon. 13 : 60, worker, $ (Type locality : Afiamalu, Upolu, 

DISTRIBUTION : SAMOA : This species is known only from Upolu, where it was 
taken in the vicinity of Afiamalu by Zimmerman in the 1940's and by the Taylors in 1962. 
Detailed collection records are given by Taylor (1967). P. loi is a member of the P. 
taipingensis species group which includes 4 other species : taipingensis Forel (Malaya), sys- 
cena Wilson (New Guinea), colaensis Mann (Fiji), and woodwardi Taylor (Upolu, Samoa). 
Among these forms syscena is the species most similar to loi. It is notable that loi was 
taken much more frequently than the sympatric woodward! by Zimmerman, while the re- 
verse was the case in the Taylor collections. It is possible that the relative abundance 
of these 2 species has changed over the last 20 years (Taylor 1967). 

Fig. 10. Ponera loi Taylor, holotype worker from Afiamalu, Upolu. (After 
Taylor, 1967.) 

1967 Wilson & Taylor : Ants of Polynesia 25 

Ponera swezeyi (Wheeler) Fig. 11. 

Pseudocryptopone swezeyi Wheeler, 1933, Amer. Mus. Novit. 672: 16-17, fig. 6, worker, 

queen (Type locality: vicinity of Honolulu, Hawaii). 
Ponera swezeyi: Wilson, 1957, Bull. Mus. Comp. ZooL, Harv. 116: 370. 

DISTRIBUTION: SAMOA: Taylor (1967) gives detailed listings of all known collec- 
tions. P. swezeyi is present on all 3 main islands : Savai'i, Upolu and Tutuila. The spe- 
cies was first collected on Upolu and Tutuila by Woodward in 1956. HAWAII : All 
known records are from Oahu, where the earliest known collections date to 1933 (see 
Taylor loc. cit.}. 

Fig. 11. Ponera swezeyi (Wheeler), syntype worker from near Honolulu, 
Oahu (After Taylor 1967.) 

The discovery in 1956 of P. swezeyi on Samoa robbed it of its apparent endemic status 
in the Hawaiian fauna. The Hawaiian records were all made from the soil of cultivated 
and fallow sugar cane fields. Taylor (1967), gives no firm statement as to the likelihood 
that this species is native on Samoa, but points out that its predilection to soil nesting 
rather than rotting logs (as usual in Ponera} may have increased the likelihood of its dis- 
persal by man. 

Ponera tenuis (Emery) 

Fig. 12. 

Cryptopone tenuis Emery 1900, Termeszetr. Fiiz. 23: 321-322, pi. 8, figs. 21, 22, worker 

(Original localities: Lemien and Tamara I., New Guinea). 
Pseudocryptopone tenuis: Wheeler, 1933, Amer. Mus. Novit. 672: 13-14. 
Ponera tenuis: Wilson, 1957, Bull. Mus. Comp. ZooL, Harv. 116 (6) : 373, fig. 2. 

DISTRIBUTION : SAMOA : Known only from Savai'i and Upolu. P. tenuis was first 
taken in Samoa, on Upolu, by Woodward in 1956. The first Savai'i records were made 
in 1962 by Taylor and Ettershank. Detailed collection records are given by Taylor (1967). 

Outside of Samoa this species is known only from several localities in NE New Guinea ; 
its occurrence in eastern Melanesia can be reasonably predicted. P. tenuis is widespread 
in rain forest at all elevations on Savai'i and Upolu. It is peculiar in Ponera as the only 
species which is known to lack pupal cocoons (Taylor loc. cit.'). 

Some aspects of the biology of this species, relative to the sympatric P. incerta, have 

Pac. Ins. Mon. 

Fig. 12. Ponera tenuis (Emery), worker from Afiamalu, Upolu. 
already been described under the latter species. 

Ponera woodward! Taylor 

Fig. 13. 

Ponera woodwardi Taylor, 1967, Pac. Ins. Mon. 13 : 62. 

DISTRIBUTION: SAMOA: Upolu [see Taylor (1966) for detailed records]. 

This species, like the somewhat similar and related P. lot, is known only from the vi- 
cinity of Afiamalu, in the center of Upolu. It was taken extensively at this locality by 
the Taylors, in disturbed and virgin rain forest, nesting in logs or under moss on tree 
trunks. It was not encountered elsewhere on the island, and may have been replaced by 
the introduced Hypoponera confinis at lower elevations. It seems likely that P. woodwardi, 
along with P. lot, is a Samoan endemic. 

Fig. 13. Ponera woodwardi Taylor, holotype worker from Afiamalu, Upolu. (After Tay- 
lor, 1967). 

Hypoponera confinis (Roger), New Combination 

Fig. 14. 

Ponera confinis Roger, 1860, Berl. Ent. Zs. 4: 284, worker (Type locality: Ceylon). 
Ponera trigona convexiuscula var nautamm Santschi, 1920, Bull. Soc. Vaud. Sci. Nat. 53 : 

Wilson & Taylor : Ants of Polynesia 

180, worker, queen (Type locality: Apia, Samoa). New Synonymy. 
Ponera trigona opacior: Wheeler, 1923, Nat. Hist. Juan Fernandez & Easter I. 3: 317. 

DISTRIBUTION : SAMOA : Savai'i : Alagaogao, ca. 100 m, several colonies from 
moss covered rotting logs in banana and cocoa plantations 20.111.1962 (RWT aces. 413- 
414, 419-422). Asau, ca. 170 m, disturbed lowland rain forest, nests from rotting logs, 
22.111.1962 (RWT aces. 493-497, 499). Gagaifoumauga, rain forest, nest in rotting log, 14. 
V.1962 (GE). Lotogo, ca. 340 m, rain forest, colony in rotting log, 21.111.1962 (RWT 
ace. 452). Matega, ca. 65 m, banana clearing at edge of rain forest, colonies under moss 
and bark on rotting logs, 20.111.1962 (RWT aces. 394, 403, 405, 408). Mt. Matafa, ca. 700 
m, rain forest, nests from rotting logs and under small epiphytes on tree trunks, strays from 
leaf mold, 23. III. 1962 (RWT aces. 520, 527-529, 532, 536, 540-543, 546). Upolu: Afia- 
malu, 700 m, beating, 1940 (ECZ). Apia, nautarum nidotypes, ca. 1920 (H. Swale) ; **1927. 
Fasitooufa, coconut plantation, nest under bark on rotting log, 29. III. 1962 (RWT ace. 605). 
Mafa Pass, disturbed rain forest, strays in leaf mold, 29.111.1962 (RWT ace. 635). Ma- 
tautu, berlesate of ground moss under Ficus, 2.II.1956 (TEW). Poutasi, rain forest, berlesate 
of leaf mold, 2.II.1956 (TEW). Sa'agafou, 
lowland rain forest, strays under moss on /^ \~^1 -^ 

logs, 29. III. 1962 (RWT ace. 609). Above f' \\ \ \\ f ~~~ 

Sulani Falls, rain forest, berlesate of ground /' ^^-^^A V *-\i \ I 
moss, 11.1956 (TEW). Tutuila : Pago Pago, | \T Yy- 

workers and <?<?, 6.VIII.1940 (ECZ) ; res- \ ' ' < V^"< """ 

ervoir at Fangatanga, 1938 (ECZ) ; Mt. '4 \^ 

Pioa, 300m, 1940 (OHS). TONGA: Eua : 

Ohonua, dealate queens, 1956 (NK). *SO- 

CIETY IS. : Tahiti : Papara Vail., 250 m, f \ / \ / i 

1925 (record based on queens, requiring ver- x" ^ I \ \\ 1 / 

ification). *E ASTER I. : ca. 1920. Y^X-^/ \ \ : ~\\. )J 

H. confinis is a widespread species which ' V \\ _ 

is probably native throughout India, Cey- \ \ v 

Ion, Indo-China, Indonesia and Melanesia, 
at least as far east as New Guinea. It is 
known from the New Hebrides but has not X" ^~~ 

been reported from the Solomons or Fiji. ^J_^-^ J \ ' \ I \ 

The Polynesian populations are probably de- [ N^r\\ ' ; *^ ' ^ ' ^^ 

rived from propagules introduced by man. _, V. ^ v , v . ^ 

This species was taken extensively by \ 

Taylor on Upolu and Savai'i, but was not 

j rr. -, . , _ Fig. 14. Hypoponera confinis fRoger). work- 

encountered on Tutuila, either by Taylor e r from Apia Upolu. Fig. 15. Hypoponera 
or Ettershank. It was also notably absent opadceps (Mayr), worker from Oahu. Fig. 
from collections at Afiamalu in central Upo- 16- Hypoponera punctatissima (Roger), work- 
lu where it may have been competitively er from Upolu ' A11 drawn to the same scale - 
excluded by the presence of Ponera woodwardi and P. lot, which seem to resemble it eco- 
logically, and which were taken only in the Afiamalu area. Most colonies contained lar- 
vae, but only one cocoon-enclosed immature (a pharate pupa) was observed. Many of 
the workers taken in colonies were callow, and in several cases $$ were present. 

Pac. Ins. Mon. 

Hypoponera opaciceps (Mayr) Fig. 15. 

Ponera opaciceps Mayr, 1887, Verb. Zool.-bot. Ges. Wien 37 : 536, worker, queen (Type 

locality: Sta. Catharina Prov., Brazil). 
Ponera perkinst Forel, 1899, Fauna Hawaiiensis (Cambridge), p. 117, worker, queen, # 

(Type locality: Hawaiian Is., in mountains, 650-1300 m). New Synonymy 
Ponera andrei Emery, 1900, Term6szetr. Ftiz. 23: 318, nota, pi. 8, fig. 47, worker (Type 

locality: Noumea, New Caledonia). 
Hypoponera opaciceps: Taylor, 1967, Pac. Ins. Mon. 13: 11. 

DISTRIBUTION: SOCIETY IS.: ^Tahiti: Papara (1927) ; Maara (1926) ; Arihiri, Pare 
(1934) ; Blue Lagoon, Papeete (1934) ; Vall6e de la Reine. Moorea : Cheesman & Craw- 
ley (1928) record opaciceps near Papeete in 1925. *AUSTRAL IS. : Raivavae : E. slope of 
Mt. Muanui, 150-200 m (1934). Rapa : numerous localities over island, from 0-500 m ; col- 
lections by E. C. Zimmerman (Wheeler 1936). MARQUESAS: Fatu Hiva (Cheesman & 
Crawley, 1928) collection made in 1928. Nuku Hiva (ibid.). *Mohotane, 300m. HAWAII: 
Perkins (in Forel 1899) found this species "in the mountains on all the islands, general- 
ly from 2,000 to 4,000 feet." Wheeler (1934) records it from collections made in the 
1920's and early 1930's from the following islands : Oahu, Molokai, Maui, Hawaii. Carter 
(1936) records a nuptial flight at Helemano, Oahu, on 1.VII.1935. 

In the New World this species ranges more or less continuously from the southern Unit- 
ed States to as far south as Montevideo, Uruguay. It is also widespread in the Old 
World, but here it is irregularly and often locally distributed, and generally scarcer (ex- 
cept for parts of Polynesia). We have studied material from Formosa, the Philippines, 
and New Caledonia (andrei type), as well as the Polynesian localities listed above. There 
is no apparently significant geographical variation in the worker caste over this vast 
range. On distributional evidence it appears that opaciceps has been introduced irregul- 
arly by commerce from the New World into the Old. 

Like the following species, P. punctatissima, opaciceps produces ergatoid males. In this 
case, however, these are much less extremely worker-like, and normal winged males are 
also produced. The details of male production in the colonies, the behavior of the erga- 
toids, and the adaptive significance of this peculiar male dimorphism remain unstudied. 

Hypoponera punctatissima (Roger) Fig. 16. 

Ponera punctatissima Roger, 1859, Berl. Ent. Zs. 3 : 254, plate 7, fig. 7, worker, queen. 

Ponera androgyna Roger, 1859, Ibid., p. 254, ergatoid ft. 

Ponera punctatissima subsp. schauinslandi Emery, 1899, Zool. Jahrb. (Syst.) 12: 439, queen 
(Type locality: Laysan I.; Syntype examined, Emery Coll.). New Synonymy. 

Ponera gleadowi: Wilson, 1958, Bull. Mus. Comp. Zool., Harv. 119: 328, taxonomy, dis- 
tribution. Taylor & Wilson, 1961, Psyche 68 (4) : 137. [Nee Ponera gleadowdi, Fore], 
1895, Mem. Accad. Sci., Bologna 5 (5) : 292.] 

Hypoponera punctatissima : Taylor, 1967, Pac. Ins. Mon. 13 : 12. 

DISTRIBUTION : SAMOA : Savai'i : Matega, 65 m. banana clearing, edge of rain forest, 

strays under bark, rotting log (RWT ace. 399). Mt. Matafa, 700m, rain forest, strays in 

rotting tree stump (RWT ace. 546). Upolu : Afiamalu, 700m, beating shrubbery, 1940 

(ECZ) ; disturbed rain forest, tree moss berlesate, workers and ergatoid $ (RWT ace. 

258). Falepuna, leaf litter, edge of swamp, 31.1.1956 (TEW). Lauli'i, 17.1.1956 (TEW). 

Wilson & Taylor: Ants of Polynesia 

Safato, Taro clearing, edge of rain forest, colony in rotting log (RWT ace. 614). TONGA: 
Eua: Ohonua, 1956 (NK). KERMADEC IS.: Raoul, Meyer, records dated VI.1956 
(Taylor & Wilson 1961). * AUSTRAL IS. : Maria : NE islet (1934). Marotiri: SE islet 
(1934). Rapa: Maitua, 200-230 m (1934). MARQUESAS: Ua-Pu: Summit of Kohepu 
(mumfordi syntypes ; 1931). HAWAII: H. punctatissima apparently occurs throughout 
the islands. Forel first recorded it from Oahu and Kauai in 1899. In Wheeler's faunal 
list of Hawaiian ants (1934) it was recorded from Oahu, Hawaii, Maui, Molokai, French 
Frigates Shoal, Necker I., and Laysan I. ; all of these records were accumulated during 
the late 1920's and early 1930's. We have been able to verify several of Wheeler's de- 
terminations of material from Oahu. 

H. punctatissima is the nominate species of a taxonomically confused group, containing 
a large number of synonymous names. Recent unpublished studies by Taylor suggest that 
only two valid species, punctatissima and gleadowi Forel, are involved. Gleadowi is un- 
doubtedly distinct from punctatissima, but the present concept of the latter could conceiv- 
ably include 2 or even 3 valid species. This does not seem very likely, but more study 
material is needed for a final analysis, and until this is in hand the name punctatissima is 
tentatively applied to the Polynesian form. The various species synonymized under glea- 
dowi by Wilson (1958) belong here, and should be provisionally transferred to the syno- 
nymy of punctatissima (these include : P. kalakauae Forel, P. gleadowi v. decipiens Forel, 
P. mina Wheeler, and P. mumfordi, Wheeler). The Polynesian material shows some vari- 
ation, but only one species seems to be present in the islands. 

H. punctatissima is virtually cosmopolitan in the warmer parts of the globe, and has 
undoubtedly been carried extensively and often by man. Gleadowi is almost as widespread, 
but it is presently unknown from the Papuasian and Polynesian areas. Both species ap- 
pear to be of Old World, probably African, origin. The known males of both are pecu- 
liar, highly worker-like ergatoids, and it appears that normal winged males are never pro- 
duced (Forel originally described the gleadowi male as winged, but his specimens were not 
collected with the worker types ; indeed, they were from the opposite side of the Indian 
sub-continent). Taylor (1967) has suggested that behavioral adaptations involving these 
males may have contributed to the remarkable abilities which these species must have for 
transport by man, and for successful establishment when introduced into new areas. 

Though widespread, punctatissima is not especially abundant in Polynesia. It is encoun- 
tered occasionally in rotting logs at forest fringes or in other disturbed but shaded situa- 
tions. The colonies are very populous and often include large numbers of dealate females. 
Virtually nothing is known of the male behavior, a subject deserving much attention. 
This is the one ponerine ant that surpasses Odontomachus simillimus and Hypoponera con- 
finis in the extent of its range in the Pacific. 

Hypoponera zwaluwenburgi (Wheeler) Fig. 17. 

Pseudocryptopone zwaluwenburgi Wheeler, 1933, Amer. Mus. Novit. 672 : 14-16, fig. 5, work- 
er (Type locality: Oahu). 

Ponera zwaluwenburgi: Wilson, 1957, Bull. Mus. Comp. Zool. Harv. 116: 375. 
Hypoponera zwaluwenburgi: Taylor, 1967, Pac. Ins. Mon. 13: 12. 

DISTRIBUTION: HAWAII: Oahu (syntype workers). Wheeler (1933) also records 
this species from Maui. 

Pac. Ins. Mon. 

Fig. 17. Hypoponera zwaluwenburgi Wheeler, syntype worker from near Hon- 
olulu, Oahu. 

This species is known only from a few worker specimens, and it is one of the few 
Hawaiian species which has not been collected elsewhere. Nonetheless we consider zwa- 
luwenburgi to be a human introduction into Hawaii. Its general features are somewhat ab- 
berant for Hypoponera, but its affinities almost certainly lie with sections of the African 
fauna. Its Hawaiian propagule probably originated in Africa. 

Leptogenys falcigera Roger 

Fig. 18. 

Leptogenys falcigera Roger, 1861, Berl. Ent. Zs. 5: 42, worker (Type locality: Ceylon). 

Leptogenys insularis Fr. Smith, 1897, J. Linn. Soc. Lond., Zool. 14: 675, worker (Type lo- 
cality: Hawaii). New Synonymy. 

Leptogenys falcigera insularis: Wheeler, 1934, Occ. Pap. B. P. Bishop Mus. 10 (21) : 11; 
1935. Ibid., 11 (11): 14. 

Fig. 18. Leptogenys falcigera Roger, worker from Oahu. 

DISTRIBUTION: LINE IS.: Washington: "Coconut" (L. A. Whitney). HAWAII: 
Wheeler (1934) lists records from localities on Kauai, Oahu, Molokai and Maui. 

L. falcigera is an African species which has undoubtedly been spread by commerce. W. 
L. Brown, Jr., has verified the above synonymy by direct comparison of type specimens. 
He also considered maxillosa Smith as a possible senior synonym of falcigera, but conclud- 
ed that it represents a second widespread African-based tropical species, with falcata 
Roger as its junior synonym. According to Brown's notes falcigera workers have notice- 

1967 Wilson & Taylor : Ants of Polynesia 31 

ably longer and broader petiolar nodes than those of maxillosa, and the males are less 

shining, more pruinose and more darkly colored. It is possible that the gap separating 

these forms may be bridged by future African accessions, but in the meantime it seems 
best to recognise them as two related species. 

Odontomachus simillimus Fr. Smith Fig. 19. 

Odontomachus simillimus Fr. Smith, 1858, Cat. Hym. Brit. Mus. 6 : 80, pi. 5, figs. 8, 9, 

queen (Original localities: Fiji Is., Ceylon). 
Odontomachus haematodes : div. auct. (nee Formica haematoda Linn6, 1758, Syst. Nat., 10th 

ed. 1: 582). 
Odontomachus haematodus var. fuscipennis Forel, 1913, Zool. Jahrb. Syst. 36: 19, worker, 

queen, $ (Original localities: Peradeniya, Ceylon; Bahsoemboe, Sumatra). 

Fig. 19. Odontomachus simillimus Fr. Smith, worker from Papua. 

DISTRIBUTION: ONTONG JAVA: Leuaniua, <?, 29.IX.1953 (J. D. Bradley). SAMOA: 
Savai'i: **Rain forest, 600-1200 m (1924). Asau, 170m, disturbed lowland rain forest, 
strays from ground, rotting log and low foliage (RWT aces. 487, 501, 504). Aopo, 170m, 
rain forest, berlesate of soil and leaf mold (RWT ace. 597). **Fagamolo (1915). Falea- 
lupo, rain forest, nest in hollow twig on ground (RWT ace. 444). Ga'utavai, coastal co- 
conut plantation, strays from palm trunks (RWT aces. 455, 456, 474) ; strays on Ficus 
branches 35 m from ground level (RWT ace. 459). Matega, 65m, banana clearing in rain 
forest, strays from moss and leaf mold on boulders (RWT ace. 390). Mt. Matafa, 700m, 
rain forest, rotting log (RWT ace. 532). Mt. Matafa Road, 200 m, rainforest, nest under 
fallen epiphyte, strays from under bark rotten log (RWT aces. 363, 367). Mt. Olomanu, 
rain forest, berlesate of moss from rotting log (GE ace. S9). Palauli (NK, 1955). Papa, 
coastal banana plantation, strays under bark on dry dadap log (RWT ace. 376). Pata- 
mea (NK, 1955). Puapua (NK, 1955). Samalaeulu, coastal rain forest, strays under 
small trunk epiphytes from rotting log, and from berlesate of leaf mold and soil (RWT 
aces. 510, 519,596). **Safune, lowlands to 300m (1917). Sili (NK, 1955). Upolu: 
Afiamalu, 700m (OHS, 1940); $$ at light, 10.VIII.1940 (ECZ). Apia, **1925 ; 1940 
(OHS) ; 1952 (J. L. Gressitt). Falepuna, berlesate of leaf litter, edge of swamp (TEW, 

"", Pac. Ins. Mon. 14 

1956). **Leulumoega (1923). Mafa Pass, berlesate of soil and moss (TEW, 1956). Ma- 
fa Pass Road (NK, 1955). **Mulifanua (1923). Siumu (NK, 1955). Tapatapao, 300m, 
under dead bark, 1955 (NK). Tanumalala, rain forest, berlesate of leaf mold (TEW). Ti- 
avi, 700m (OHS, 1940). Togitogiga, 400m, rain forest, berlesate of leaf litter (GE aces. 
56, 57). Utumapu (TEW, 1956). Tutuila: Amouli, 120-250 m, beating shrubbery, $, 2. 
VIII.1940 (ECZ) ; **1923. **Fagasa (1923). Fagatoga Trail, 1938 (ECZ) ; 1952 (C. P. 
Hoyt). **Leone Road (1921). Matafao Trail, 400 m, beating dead branches (ECZ, 1940). 
Mt. Piao, 300m, 1940 (OHS). Between Oloava and Olomanu, 430m, rain forest, berle- 
sate, moss ex fallen log (GE ace. T15). Pago Pago, 100 m, 1910 (M. Emery, 1914). 
**Eastern end of island, 350m (1918). Manua: Ta'u, **1923 ; 1926 (A. F. Judd). TON- 
GA: Tongatabu: [W. Cottrell-Darmer, record in Emery (1914a)]. Nukualofa (NK, 1956). 
Haamonga (NK, 1956). Vavau : Neiafu (NK, 1956). Mangia (NK, 1956). ROTUMA : 
(All collections by H. St. John, 3-15.VIII.1938). Jarua. Kilinga, near Kongai, including ^ 
collected 13.VIII. at Lulu Bay. Paho. Solmatiha. Soloroa. TOKELAU IS.: Fakaofo: 1924 
(EHB). SOCIETY IS: Tahiti: Fautaua Valley, 23 m, worker (date ?-in Wheeler 1936). 
Arihiri, Pare, $, 26.111.1934 (ECZ). 

O. simillimus or a closely related species has been intercepted at the U. S. Dept. of 
Agric. Quarantine Station at Honolulu (Wheeler 1934a), but no member of the genus has 
become established in the Hawaiian Is. 

As noted in Wilson's recent revision of the Melanesian Odontomachini (Wilson 1959), 
O. simillimus is a native Indo-Australian species distinct from the Neotropical O. haemato- 
dus. It ranges continuously from Ceylon to Micronesia and inner Polynesia. Within this 
range it is one of the dominant ants, occurring in nearly every island and maintaining 
dense populations in a wide variety of habitats. Wheeler's 2 records of the species from 
Tahiti have not been verified during the present study. If simillimus does indeed oc- 
cur on that island, it must be very rare. In fact, the population was almost certainly in- 
troduced by man from its native range in the western Pacific and possibly only very re- 
cently. It is significant that not a single collection was made by Miss Cheesman during 
intensive collecting on Tahiti in 1925, despite the fact that the ant is large and conspicu- 
ous and usually one of the first to be picked up by collectors. 

O. simillimus shows no detectable geographic variation in any of the castes over its 
great range. It is most closely related to a group of Australian and Papuan species that 
includes aciculatus Fr. Smith, aeneus Emery, and cephalotes Fr. Smith. 

Anochetus graeffei Mayr 

Fig. 20. 

Anochetus graeffei Mayr, 1870, Verh. Zool.-bot. Ges. Wien 20: 961, worker (Type locality : 
Upolu, Samoa). Wilson, 1959, Bull. Mus. Comp. Zool., Harv. 120: 507, distribution, 
variation, biology. 
Anochetus puncti\entris subsp. oceanicus Emery, 1897, Term6zetr. Fuz. 20: 597, worker (Type 

locality : Friedrich-Wilhelmshafen, New Guinea). 

DISTRIBUTION: SAMOA: Savai'i: Alagaogao, 100m, banana plantation, strays from 
rotting wood fragments on ground (RWT ace. 417). Upolu : Afiamalu, 700m, beating 
(ECZ, 1940). Matautu, from Berlese funnel sample of leaf mold (TEW, 1956), Tutuila: 
Alega, coastal Futu grove Berlese funnel sample (TEW, 1956). **Amouli (1923). Naval 

Wilson & Taylor : Ants of Polynesia 


Fig. 20. Anochetus graeffei Mayr, worker from Afiamalu, Upolu. 

Station, $&, 15.VIII.1940 (OHS, ECZ). Utulei, 15m (ECZ, 1940). Manila Group: Ta'u 
(1937). COOK IS. : Rarotonga, 1937 (C. E. Clarke). 

A. graeffei ranges from SE Asia to Queensland, and eastward into Micronesia and the 
Cook Is. It is continuously distributed throughout Melanesia, including New Caledonia. 
Striking non-geographic variation is shown by the worker caste, in total size, petiole shape 
and sculpturation, especially that of the 1st gastric tergite (see Wilson 1959). The species 
is rather scarce in Samoa. 


Syscia silvestrii Wheeler Fig. 21. 

Cerapachys (Syscia') silvestrii Wheeler, 1909, Boll. Lab. Zool. Gen. Agrar. Portici 3 : 269 

(Type locality: Hilo, Hawaii). 

Cerapachys (Syscia') seini Mann, 1931, J. Wash. Acad. Sci. 21: 440-441, worker (Type lo- 
cality: Rio Piedras, Puerto Rico). New Synonymy. 

uli'i. Salea Puga. These 2 records, which ^%X >^t\j'j|T >t 

include a large number of specimens, were Jr /"M\x- > ^^^*<Cr^~~ '^ 

both made by Woodward during January ^ Wi J^ v - s * v - sJS - S1 ^ ^% 

1956. Both are from Berlese funnel sam- Q'^v o X 

pies. S. silvestrii was not taken by the ^^> ^ 

Taylors or Ettershank, and no earlier re- Fjg 21 Syscia typMa Roger> worker from 

cords are known. HAWAII : Hawaii : Hilo Salea Puga, Upolu. 

Pac. Ins. Mon. 14 

(type locality). Oahu: (R. H. Van Zwaluwenburg) . Wheeler (1934) also recorded silves- 
trii from Honolulu. 

S. silvestrii is probably a pantropical tramp species, although certainly one of the most 
inconspicuous and seldom collected of this class. Besides the Samoan, Hawaiian, and syn- 
onymic type material cited above we have seen workers from Okinawa (F. G. Werner) 
and Amlekhganj, Nepal (E. I. Coher). The series show very little variation in the char- 
acters of size, body form, and sculpturing that often provide diagnostic species-level tax- 
onomic characters in Syscia. Considerable internidal variation does occur in the depth of 
pigmentation and shininess of interpunctural cuticle, but its significance is uncertain, since 
these are the features most likely to be modified by preservation. 

Cerapachys (Syscia) sinensis Wheeler (1928, Boll. Lab. Zool. Gen. Agrar. Portici 22: 3, 
worker, Shanghai, China) is probably a synonym of silvestrii, but we are reluctant to for- 
mally propose this on the basis of the very few specimens available for comparison. 

We have not seen the type of the related S. typhla Roger, originally described from Cey- 
lon. It was examined recently by Dr W. L. Brown, Jr., who reports that typhla is a spe- 
cies distinct from silvestrii. It has a longer head, coarser sculpturation, and the postpetiole 
is relatively very long, being distinctly more than half as long as the 1st gastric segment. 
In other Syscia species the postpetiole is much less than half as long as gastric segment 
I ; usually only about one third its length. 

Eurhopalothrix procera (Emery) Fig. 22. 

For complete current synonymy see Brown & Kempf (1960). 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, beating, 1940 (ECZ) ; ground 
moss berlesates, disturbed rain forest, 1.1956 (TEW) ; (RWT aces. 2316, 2321) ; tree moss 
berlesates, ca. 8 m from ground, disturbed rain forest, 9.IV.1962 (RWT ace. 580 ; GE ace. 
44) ; strays ex small rotting wood fragments, disturbed rain forest (RWT aces. 274, 353) ; 
nest ex rotting log, disturbed rain forest, 19.111.1962 (RWT aces. 218, 227) ; nest ex rot- 
ting stump, rain forest, 16.111.1962 (RWT ace. 329). **Malololelei (1924, 1926). Tiavi, 
700 m, under dead bark, 1940 (ECZ). 

Brown & Kempf (1960) find this to be the most widely^distributed and variable of the 
Indo-Australian Eurhopalothrix. It ranges from Sarawak and Koto Sho I. (Bolel'Tobago), 

Fig. 22. Eurhopalothrix procera (Emery), worker from Afiamalu, Upolu 

1967 Wilson & Taylor : Ants of Polynesia 35 

SE of Formosa, through the Philippines and Moluccas to Cape York, Palau, Santa Cruz 
Is., and Samoa. Notable geographical variation occurs in size, coarseness and opacity of 
body sculpturing, and degree of reduction of pilosity. In general, samples from peripheral 
areas are larger, more coarsely sculptured, and have more complete pilosity than samples 
from the center of the range in New Guinea proper. The Samoan workers conform to 
this pattern, except that the sculpturing is reduced to an intermediate degree ; in this lat- 
ter character they resemble workers from Cape York. For further details, the Brown & 
Kempf revision should be consulted. 

E. procera was present but not abundant at Afiamalu in 1962. Only 2 colonies were 
found, 1 in a rotting log and the other in a rotting tree stump in disturbed rain forest. 
Each contained alate forms of both sexes as well as several hundred workers, together 
with large numbers of larvae and pupae. 

Trichoscapa membranifera Emery 

Fig. 24. 

Strumigenys (Trichoscapa) membranifera Emery, 1869, Ann. Acad. Asp. Nat. Napoli 2: 24- 

25, pi. 1, worker. 
Strumigenys (Cephaloxys) membranifera var. williamsi Wheeler, 1933, Proc. Hawn. Ent. 

Soc. 8: 275, fig. 1, worker (Type locality: near Olaa, Hawaii). 
Trichoscapa membranifera: Brown, 1949, Mushi^lO (1): 6. 


Fig. 23. Quadristruma emmae (Emery), worker from Lauli'i, Upolu (TEW). Fig. 24. 
Trichoscapa membranifera Emery, worker from Vaipoto, Upolu (TEW). Fig. 25. Smithi- 
struma dubia Brown, worker from Lauli'i, Upolu (TEW). 

DISTRIBUTION: SAMOA: Upolu: Apia, $ at light, 4-9.IV.1962 (GE). Vaipoto, Ber- 
lese funnel sample, leaf mold, rain forest, 9.1.1956 (TEW). HAWAII: Hawaii: off road 
to Puna, near Olaa (williamsi syntypes). 

Brown (1949a) states that T. membranifera is probably of African origin. It has been 
spread by human commerce through a large part of the tropics and warm temperate zones, 
including such diverse areas as the Fiji Is., eastern China, West Indies and southeastern 
United States. The species has an ecological amplitude unusual for a dacetine, nesting 
in major habitats from dense woodland to dry, open cultivated fields. Wilson (1953) de- 
scribes the feeding behavior of the workers ; they are predaceous on a wide variety of 
small, soft-bodied arthropods. 

Pac. Ins. Mon. 

Smithistruma dubia Brown 

Fig. 25. 

Smithistruma (Smithistruma') dubia Brown, 1953, Amer. Midi. Nat. 50: 122, pi. 1, fig. 13, 
worker, gynetype . (Original localities: in Palau). 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, berlesate of tree moss, 2 m above 
ground, 11.111.1962 (RWT ace. 262) ; berlesate, ground moss, disturbed rain forest, 9.IV. 
1962 (GE ace. 53). Lauli'i, berlesate, leaf litter and soil, rain forest, 8.1.1953 (TEW). 

When Brown described this species it was listed only from the islands of Peleliu and 
Garakayo of the Micronesian Palau group. At that time it was suggested that dubia was 
probably a tramp, originating in the large Papuan or Indonesian land masses to the south 
and west. More recent Melanesian and Micronesian records published by Brown (1964) 
support this hypothesis. These records include the following : Caroline Is. : Truk, Ponape ; 
NE New Guinea : Bubia, Lower Busu River ; Papua : Bisianumu. In addition we have 
seen material from the Solomon Is. of Guadalcanal and Bellona, and from Reef I. in the 
Santa Cruz group (all collected by P. J. M. Greenslade). S. dubia has not been taken on 
the Fiji Is., but probably occurs there. There seems little doubt that this species is New 
Guinea-based, and it probably spreads by commerce. 

Smithistruma mumfordi (Wheeler) 

Strumigenys (Cephaloxys) inezi [!] subsp. mumfordi Wheeler, 1932, Bull. B. P. Bishop Mus. 

95 : 160, worker [Type locality (by restriction of W. L. Brown, 1953, Amer. Midi. Nat., 

50: 124) : Hakahetu Vail., Ua Pu, Marquesas]. 
Smithistruma mumfordi : Brown, 1953, Amer. Midi. Nat. 50 : 124. 

DISTRIBUTION: MARQUESAS: Ua Pu: Hakahetu Vail., 800m (A. M. Adamson). 
Nuku Hiva: Ooumu, 1300m (Mumford & Adamson). 

W. L. Brown, Jr. (1964) has recently redescribed a syntype of this species, and has been 
able to clarify some of the difficulties engendered by Wheeler's inadequate original de- 
scription (see Brown 1954). 

S. mumfordi is a member of the Oriental-Papuan capitata group, which contains at least 
8 other species. In Brown's 1954 key to the group, it runs to the couplet including pedun- 
culata Brown of the Philippines and capitata (Fr. Smith) of western New Guinea. It is 
close to these two species and also to S. inezae Forel, with which it could be synonymous. 
The species was probably introduced by man into the Marquesas from Indonesia or Mela- 

No information is available on the feeding behavior of this or any other Polynesian 
species of Smithistruma or Strumigenys. Elsewhere these genera are predominantly preda- 
ceous on certain groups of collembolans and a few other small, soft -bodied arthropods (Brown 
& Wilson 1959). 

Strumigenys godeffroyi Mayr 

Fig. 26. 

Strumigenys godeffroyi Mayr, 1866, Sitz. Akad. Wiss. Wien 53 : 516, worker. Brown, 1949, 

Mushi 20: 17, synonymy; 1959, Psyche 65: 81, distribution. 

DISTRIBUTION : SAMOA : Savai'i : Gagaifoumauga, berlesate, moss and bark on rotten 
log, rain forest, 14. VI. 1962 (GE ace. 55). Falealupo, several nests in rotten logs, rain forest, 
20.111.1962 (RWT aces. 434, 435). Lotogo, 370 m, nests ex rotten logs, rain forest, 21.111. 

Wilson & Taylor: Ants of Polynesia 

Fig. 26. Strumigenys godeffroyi Mayr, worker from Colonia, Ponape I., Caroline 

1962 (RWT aces. 464, 465). Mt Matafa Road, 200 m, rain forest edge, strays under bark 
rotten log, 19.111.1962 (RWT ace. 365). Upolu : Afiamalu, 700m, beating, 1940 (ECZ) ; 
berlesate, moss and ferns off trees, rain forest, 19.1.1956 (TEW) ; berlesate, moss on tree 
trunks l-2m above ground level, rain forest, 11. III. 1962 (RWT ace. 262) ; nest ex rotten 
log, Taro patch, 16.111.1962 (RWT ace. 320) ; Apia (1924). Lauli'i, berlesate, leaf litter 
and soil, rain forest, 8.1.1956 (TEW). Malololelei, 650m, berlesate, rain forest, 19.1.1956 
(TEW). Sinaele, 500m, beating, 1940 (ECZ). Tanumalala, 350m, strays under moss on 
buttress roots and fallen branch (RWT aces. 606, 608). Tapatapao, 300m, under dead 
bark, 1940 (ECZ). Tiavi, 700m, rotting logs, 1940 (ECZ). Utumapu, 120m, 17.1.1956 
(TEW). This species was originally described from specimens originating from Upolu 
sometime before 1866. Tutuila : North of Aloau, 450 m, berlesate, moss ex logs, stunted 
forest, 13.VII.1962 (GE ace. T. 19). Malaeimi, berlesate, moss ex rotten logs, rain forest, 
13.VII.1962 (GE ace. T. 22). Matafao trail, 400m, beating dead branches, 1940 (ECZ). 
Pago Pago Matafao Trail, 400m, beating shrubbery, 1940 (ECZ). Pago Pago, north side 
300m, under dead bark, 1940 (ECZ). Between Oloava and Olomanu, 430m, berlesate, 
moss ex rotten logs, rain forest, 13.VII.1962 (GE aces. T. 12, T. 15). SOCIETY IS. : Ta- 
hiti: Papenoo Vail., 8km from sea, 300m, 1929 (Adamson). AUSTRAL IS.: Raivavae: 
Unurau (1934). HAWAII: Oahu: Poamoho, Koolau Range, under rock at margin of 
wet forest, 27.VIII.1961 (C. H. Lindroth). 

W. L. Brown, who has conducted a careful study of the Indo-Australian Strumigenys, 
considers that godeffroyi is a native of tropical southeastern Asia. It shows greater var- 
iability on Borneo and Sumatra than elsewhere, and the most closely related known species 
(indagatrix Wheeler, juliae Forel) coexist with it in the greater Sundas. The related species 
lewisi Cameron replaces godeffroyi in subtropical and warm temperate eastern Asia (Brown 
1949a). S. lewisi also replaces it for the most part as an introduced species on Hawaii. 
Brown (ms., revision of Indo-Australian Strumigenys} states: "godeffroyi is widely distributed 
in even the most far-flung Pacific Island groups and in the East Indies, New Guinea, 
Solomons, tropical mainland of Asia, Philippines, northern Australia, and across the Indian 
Ocean as far as Madagascar, where it has apparently been carried by commerce. The great 
majority of the records from larger land-masses are littoral, showing that this ant is a 
strong transoceanic traveller, thriving in areas where the natural vegetation has been re- 
placed by the tropicopolitan lowland flora associated with the works of man. Wheeler 

Pac. Ins. Mon. 

(1937, Treubia 16: 23) records that godeffroyi was found by Dammerman to have rees- 
tablished itself in Krakatoa by 1933." Except for more extensive variability of populations 
in the greater Sundas, the species appears to show no significant geographic variation over 
its extensive range (Brown 1949a). 

Strumigenys lewisi Cameron Fig. 27. 

Strumigenys lewisi Cameron, 1887, Proc. Manchester Lit. Phi. Soc. 25: 229, worker Brown 

1949, Mushi 20 (1) : 16. 
Strumigenys godeffroyi var. or subsp. lewisi of authors. 

DISTRIBUTION : *HAWAII : Oahu : Palolo Vail. Manoa, 300 m, Waihii Falls Manoa 

This species is native to parts of China, Japan, Upper Burma and the Ryukyus, where 
it replaces the related species godeffroyi (q. v.). Brown (1949a) states that the species has 
been intercepted at U. S. Plant Quarantine Stations in ginger root and has reached various 
Pacific Is. (unspecified) " in shipments of timber and in the earth about living plants." 
Brown reported having collected several nests in the densely populated, irrigated plains 
around Chengtu, Szechuan Province, China. 

S. lewisi closely resembles the much more abundant, widespread godeffroyi. As indicated 
by the accompanying figures, it is most readily separated by its much sparser pronotal 

Fig. 27. Strumigenys lewisi Cameron, head of worker from Chengtu, Szechwan, 
China and body of worker from Shantung, China. Not drawn to the same 

Strumigenys mallei Wilson and Taylor, new species Fig. 28. 

DIAGNOSIS AND RELATIONSHIPS. Similar to S. perplexa (Fr. Smith) of Australia and New 
Guinea, differing by its larger size, distinctly more slender head, proportionately larger 
mandibles, more slender propodeal spines, projecting posterior rim of postpetiolar disk, and 
overall much sparser pilosity. One way it can be distinguished immediately from perplexa 
is that less than 10 standing hairs break the dorsal profile of the mesosoma, whereas in 
perplexa over 15 do. Another similar species, S. dyak Wheeler of Borneo, can be distin- 
guished from mallei by its much deeper occipital concavity, by its smaller postpetiolar 
disk, and by its much longer, differently arranged mesosomal pilosity. 

Wilson & Taylor : Ants of Polynesia 

Fig. 28. Strumigenys mallei Wilson and Taylor, holotype worker. 

Holotype worker (SAMOA) : HW 0.49 mm ; HL 0.75 mm, SL 0.45 mm, PW 0.31 mm, 
ML 0.34 mm. Body form and pilosity as shown in fig. 28. All of head, entire dorsum 
of mesosoma, and all of petiole densely " shagreened ", i. e., microreticulate with diameter 
of interreticular spaces about 0.01 mm ; surface completely opaque in reflected light, almost 
all sides of mesosoma, exposed dorsal surface of postpetiole, and all of gaster, smooth 
and shining. Concolorous light reddish brown. 

Paratype worker (FIJI) : HW 0.49 mm, HL 0.73 mm, SL 0.43 mm, PW 0.31 mm, ML 
0.31 mm. Fringing hairs of anterior clypeal border more numerous and flattened than in 
holotype. Shagreening covers side of pronotum as well as dorsum. The 2 short, curved 
hairs found on propodeal dorsum of holotype (see fig. 28) are lacking. Otherwise this spec- 
imen is nearly identical to the holotype. 

MATERIAL EXAMINED : SAMOA : Upolu : Afiamalu, 700 m, rain forest, berlesate of moss 
on tree, 10-12 m from ground, 15.111.1962, holotype worker (RWT ace. 581). FIJI: Viti 
Levu: Navai Mill, near Nandarivatu, 800m, 17.IX.1938 (E. C. Zimmerman). 

The name is based, in apposition, on the Samoan word for " trap. " It alludes to the 
remarkable, trap-like action of the mandibles of workers of this genus. 

TYPE DEPOSITION : Holotype and paratype deposited in the Harvard University Museum 
of Comparative Zoology (Type no. 31114). 

Strumigenys rogeri Emery 

Fig. 29. 

See Brown, 1954, Bull. Mus. Comp. Zool., Harv. 112: 20, for complete synonymy. 

DISTRIBUTION : SAMOA : Savai'i: Asau, 170 m, strays ex rotten log, rain forest (RWT 
ace. 504) and berlesate of leaf mold, 22.111.1962 (RWT ace. 597). Falealupo, nest in rotten 
log, rain forest, 20.111.1962 (RWT ace. 437). Gagaifomauga, berlesates, leaf mold, soil 
and moss on rotten logs, rain forest, 14.VI.1962 (GE aces. S1-S4, S6). Lotogo, 350m, strays 
under bark rotten log, rain forest (RWT ace. 450) ; nests in rotting wood fragments (RWT 
aces. 478, 479) ; nests under stones (RWT aces. 480, 481) ; all 21.111.1962. Matega, 65m, 
banana clearing, forest edge, strays from moss and leaf mold on boulders, 20.111.1962 

Pac. Ins. Mon. 

Fig. 29. Strumigenys rogeri Emery, worker from Vaipoto, Upolu (TEW) 

(RWT aces. 390, 391). Mt. Matafa, berlesate of tree moss, 23.111.1962 (RWT ace. 588). 
Mt. Matafa Road, 200m, strays in rotten wood rain forest, 19.111.1962 (RWT ace. 364). 
Mt. Olomanu, berlesates, moss, soil and epiphytes on logs, rain forest, 15.VI.1962 (GE aces. 
S7-S11). Samalaelulu, coastal rain forest, strays under small trunk epiphytes (RWT ace. 
509) ; nest in rotten log (RWT ace. 519) ; berlesate of leaf mold and soil (RWT ace. 596); 
all 22.lll.1962 Upolu: Afiamalu, 700m, berlesate, leaf mold, rain forest, 6.1.1956 (TEW); 
nest under moss on tree trunk, 1.5m from ground level, 25.111.1962 (RWT ace. 598) ; nest 
series from small rotting wood fragments, disturbed rain forest, 9-15.III.1962 (RWT aces. 
237, 238, 304); berlesate, moss on tree trunks 1.2m from ground level, dealate and 
workers, 11. III. 1962 (RWT ace. 262) ; nests under moss on ground, partly cleared rain 
forest, 12.111.1962 (RWT aces. 228, 281) ; strays in rotten log, 9.III.1962 (RWT ace. 224) ; 
berlesates, moss on logs in disturbed rain forest, III-IV.1962 (RWT aces. 2315, 2320; GE 
aces. 24, 47, 48) ; berlesates of tree moss at 8-9 m from ground in disturbed rain forest 
(RWT aces. 2305, 2306; GE aces. 43, 45). Fagoloa, 170m, berlesate moss on log, rain 
forest, 13.VI.1962 (GE ace. 67). Falepuna, berlesate, leaf litter, edge of swamp, 31. HI. 
1956 (TEW). Above Falevao, 200m, berlesate, moss and soil, rain forest, 1.1956 (TEW). 
Lauli'i, berlesate, 17.1.1956 (TEW). Le Mafa, 400m. berlesate, soil and moss, 10.1.1956 
(TEW) ; berlesates of moss, soil and epiphytes, rainforest, 13.VI.1962 (GE aces. 61, 63, 65, 
66). Malololelei, 660m, berlesate of moss, rain forest, 19.1.1956 (TEW). Matautu, berle- 
sate, leaf mold under Ficus, 2.II.1956 (TEW). Salea Puga, berlesate, leaf mold, coastal 
Futu grove, 10.1.1956 (TEW). Sliding Rock, 200m, nests and strays under moss and on 
streamside rocks, 13.111.1962 (RWT aces. 286, 287). Tanumalala, berlesate, leaf mold, rain 
forest, 2.II.1956 (TEW). Tiapapata, berlesates of moss on rotten logs, rain forest, 19.VI. 
1962, (GE aces. 69, 71, 75). Togitogiga, 400m, berlesates, leaf litter, rain forest, 12.VL 
1962 (GE aces. 57, 58). Utumapu, 17.1.1956 (TEW) ; nests under moss and small trunk 
epiphytes on trees, up to 2m above ground level, 17.111.1962 (RWT aces. 568, 575). Tutuila: 
North of Aloau, 450 m, berlesates, moss on logs, rain forest, 13.VII.1962 (GE aces. T17, 
T19, T20). Le Mafa, 250m, berlesates of tree fern roots and moss on rock cliff, rain 

1967 Wilson & Taylor : Ants of Polynesia 41 

forest, 12.VII.1962 (GE aces. T6, T10). Malaeimi, berlesates, moss on rotten logs, rain 
forest, 13.VII.1962 (GE aces. T21-T24). Between Oloava and Olomanu, berlesates of moss, 
leaf litter and soil from rotten logs on forest floor, rain forest, 13.VII.1962 (GE aces. T13- 
T15). Tafuna, berlesate, soil and leaf mold, rain forest, 26.1.1956 (TEW). SOCIETY IS. : 
Tahiti: Pounaauia district, open field, 1949 (J. M. Dixon). HAWAII: Oahu: Tantalus, 
Honolulu, 1933 (F. X. Williams). This may be the species recorded by Williams (1937) 
as S. lewisi Cameron, from Mount Olympus Trail. 

The distribution and biology of this species have been discussed by Brown (1954). It 
is native to Africa and has been spread widely in the tropics by human commerce. 

Strumigenys szalayi Emery Fig. 30. 

Strumigenys szalayi Emery, 1897, Termeszetr. Ftiz. 20: 578, pi. 14, fig. 10, worker, $.. 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, beating shrubbery, 1940 (ECZ) ; 
berlesate of moss and ferns off trees, rain forest, 19.1.1956 (TEW) ; nest with alates ex 
rotten wood fragment, rain forest, 16.111.1962; nest in small rotten wood fragment, partly 
cleared forest, 17.111.1962; nest with alates in rotten treefern stump, Taro patch, 19.111.1962 
(all RWT aces. 318, 319, 334, 352, 603) ; berlesates, moss on logs, disturbed rain forest, 
IH-IV.1962 (RWT ace. 2315); berlesate of tree moss at llm, 9.IV.1962 (GE ace. 37). 
Malololelei Road, 590m, dead leaf sheaf of Clinostigma, 8.VII.1940 (OHS). Tiapapata, 
berlesate of epiphyte ferns, rain forest, 19.VI.1962 (GE ace. 72). Tiavi, 700m, 1940 (ECZ). 

Fig. 30. Strumigenys szalayi Emery, worker from Boana, N.-E. New Guinea 
(E. O. Wilson). 

W. L. Brown, who is currently revising the szalayi group is of the opinion (pers. comm.*) 
that the series just cited constitute a recognizable Samoan form close to szalayi Emery. 
Workers of szalayi from the lowlands of the N. coast of New Guinea are distinguished 
by small size (HL 0.68-0.77 mm) and blackish-brown gaster contrasting with ferruginous 
body. Those from the Philippines, Queensland, parts of New Guinea and the New Hebrides, 
mostly at intermediate elevations, are distinguished by larger size (HL 0.74-0.91 mm) and 
concolorous ferruginous body. The Samoan population, which is on the borderline of 
distinctness as a species (allopatric semi-species), combines quite large size (HL 0.81- 
0.90 mm) with a bicolorous body resembling that of northern New Guinean szalayi. 

Pac. Ins. Mon. 

Quadristruma emmae (Emery) Fig. 23. 

Epitritus emmae Emery, 1890, Bull. Soc. Ent. Ital. 22 : 70, pi. 8, fig. 6, worker (Type loca- 
lity: St. Thomas, Virgin Is., West Indies). 
Epitritus wheeleri Donisthorpe, 1916, Ent. Rec. 28: 121, worker (Type locality: Honolulu, 

Quadristruma emmae: Brown, 1949, Trans. Amer. Ent. Soc. 75: 47. 

DISTRIBUTION: SAMOA: Upolu: Lauli'i, Berlese funnel sample, 17.1.1956 (TEW). 
HAWAII: Oahu: Waimanalo (Swezey 1922). *Waimanalo, in cane stool. 'Honolulu. 

Brown (1949b) believes Q. emmae to be African in origin. It is related to the African 
Strumigenys rogeri group through the intergradient S. tetraphanes Brown. Emmae has been 
spread by human commerce to the following tropical areas (Brown 1949b) : Hawaii, Guam, 
Southern Florida (U.S.A.), Puerto Rico, St. Vincent, Cuba, Surinam, Sumatra, Singapore! 
New Guinea. More recently this species has been collected in the islands of Negros, 
Philippines, and Espiritu Santo, New Hebrides. It has also been taken at Kuranda, North 
Queensland by Taylor (ace. 1362, 4. VI. 1962). 

Pheidole (P.) aana Wilson and Taylor, new species Fig. 31. 

DIAGNOSIS AND RELATIONSHIPS : A medium-sized member of the variabilis group of species, 
which are mostly limited to Australia. Among these, ampla Forel, mackayensis Forel, and 
tasmaniensis Mayr differ from aana in their larger size, darker soldier coloration, more 
rounded humeri in the soldier, proportionately shorter propodeal spine in the soldier, and 
shining minor workers. Variabilis Mayr differs from aana in having the cephalic rugae 
extending into the posterior 1/2 of the head surface and in having the pronotum trans- 
versely rugose ; also, its minor worker is mostly shining. Vigilans Fr. Smith closely resem- 
bles aana but is larger, more lightly colored, and possesses a distinctively large, evenly 
convex pronotum. Yarrensis Forel is smaller, has a more rounded mesosomal profile in 
the soldier and has a shining minor worker. We have been unable to discover any Mela- 
nesian species in our extensive determined and undetermined collections that resemble 
aana as closely as these Australian species. 

Holotype soldier : HW 1.48 mm, HL 1.65 mm, SL 0.74 mm, PW 0.65 mm. Head very large 
proportionate to remainder of body. Head shape, body shape, and cephalic sculpturing as 
illustrated in fig. 31. Dorsum and anterior halves of sides of pronotum mostly smooth 
and shining. Posterior halves of sides of pronotum, entire mesothorax, and most of pro- 
podeum irregularly rugose, rugae spaced widely apart, their interspaces shagreened and 
subopaque. Anterior halves of petiolar and postpetiolar nodes smooth and shining, remaind- 
er of pedicel irregularly rugose, interspaces shagreened and subopaque. Abdomen shining. 
Head and mesosoma light, clear reddish brown, grading to clear brownish yellow in pos- 
terior 1/2 of mesosoma. Remainder of body clear brownish yellow. 

Paratype minor worker : (From holotype nest series). HW (including eyes) 0.55 mm, HL 
0.56 mm, SL 0.55 mm, PW 0.34 mm. Mesosoma and pedicel as shown in fig. 31. Mandibles 
feebly shining, striate, clypeus, antennal fossae, and entire sides of head rugoreticulate ; 
entire head surface, including reticular interspaces, " shagreened ", that is, evenly micro- 
rugoreticulate with microreticular diameters all about 0.01 mm. Entire mesosoma " sha- 
greened " in the same fashion. Head and mesosoma surfaces mostly opaque. Anterior 
dorsal surface of petiolar node smooth and shining, remainder of surface microrugoreticulate 

Wilson & Taylor : Ants of Polynesia 

Fig. 31. Pheidole aana Wilson and Taylor, holotype soldier (left and upper right), 
paratype minor worker (lower). 

and opaque. Postpetiole and abdomen mostly smooth and shining. Body concolorous 
light brown ; appendages a contrasting brownish yellow. 

Other paratypes : The other members of the holotype nest series show little deviation 
from the individuals described above. A single additional soldier has HW 1.53 mm ; a 
random sample of 10 minor workers has HW range 0.53-0.57 mm. The other nest series 
(RWT ace. 246) differs markedly in coloration : both the soldier and minor workers are 
reddish yellow. This could be an artifact of preservation. Otherwise this second series 
resembles the holotype nest series closely. The soldier has HW 1.46mm; the HW of the 
minor workers ranges 0.58-0.61 mm. 

MATERIAL EXAMINED : SAMOA: Upolu: Afiamalu, 16.111.1962, undisturbed rain forest, 
under stone, holotype soldier, 1 paratype soldier, 34 paratype minor workers (RWT ace. 
333) ; 10. III. 1962, disturbed rain forest, in rotting log, paratype soldier, 5 paratype minor 
workers (RWT ace. 246). 

The species is named in apposition after Aana, one of the four ancient districts of Samoa. 

Pheidole (P.) atua Wilson and Taylor, new species Fig. 32. 

DIAGNOSIS AND RELATIONSHIPS : Similar to P. knowlesi Mann of Fiji, differing in the 
soldier caste in having the entire dorsum of the posterior part of the head and of the 
pronotum roughly sculptured (occipital lobes and pronotum smooth and shining in knowlesi) 
and the postpetiole seen from above elliptical but not spinose (sides of postpetiole drawn 

44 Pac. Ins. Mon. 14 

out into spines in knowlesi). Differing in the minor caste in the possession of abundant, 
distinctively patterned rugae over the head and thorax. Together, knowlesi and atua are 
phenetically rather close to the variabilis group of the Indo-Australian area, including P. 
aana Wilson & Taylor of Samoa. 

Holotype soldier: HW 1.13mm, HL 1.18mm, SL 0.68mm, PW 0.48mm. Head shape, 
body shape, and cephalic sculpturing as illustrated in fig. 32. Cephalic rugal interspaces 
feebly shagreened and subopaque to feebly shining. Pronotum and mesonotum crossed 
with about 12 transverse, evenly spaced rugae ; interspaces feebly shagreened and moder- 
ately shining. Several transverse rugae on anterior 1/2 of basal face of propodeum. Most 
of propodeum, pedicel, and gaster smooth and shining. Body concolorous yellowish red, 
the appendages a shade lighter, approaching clear yellow. 

Paratype minor worker : (Chosen at random from RWT ace. 308). HW (including eyes) 
0.58 mm, HL 0.58 mm, SL 0.60 mm, PW 0.38 mm. Mesosoma and pedicel forms as shown 
in fig. 32. Sculpturing very distinctive. Head covered with over 30 fine transverse rugae 
spaced on the average about 0.02mm apart. These gradually curve to a longitudinal a- 
lignment on occiput, so that those around midline of head at occiput are nearly perfectly 
longitudinal. Scattered longitudinal cross-rugae join dominant transverse ones, creating 
a weak reticular effect in places. Interfaces feebly shagreened and subopaque. Thorax 
similarly sculptured. Propodeum lacks distinct rugae but is more heavily shagreened (mi- 
cro reticulate with reticular diameters about 0.1 mm) and subopaque. Pedicel and gaster 
smooth and shining. 

Fig. 32. Pheidole atua Wilson and Taylor, holotype soldier (left and upper 
right), paratype minor worker (lower). 

1967 Wilson & Taylor : Ants of Polynesia 

Other paratypes: Little variation is shown. Two soldiers both have HW 1.07mm. A 
random sample of 10 minor workers has HW 0.56-0.60 mm. 

MATERIAL EXAMINED : SAMOA : Upolu : Afiamalu, 700 m, partly cleared rain forest, fallen 
tree trunk, 17.111.1962, holotype (RWT ace. 353) ; disturbed rain forest, from rotting log, 
16.111.1962, 2 paratype soldiers and 25 paratype minor workers (RWT ace. 308). 

The species is named after Atua, one of the four ancient districts of Samoa. 

Pheidole (P.) fervens Fr. Smith 

Fig. 33. 

Pheidole fervens Fr. Smith, 1858, Cat. Hym. Brit. Mus. 6: 176 (Type locality: Singapore). 
Pheidole javana Mayr, 1867, Tijdschr. Ent. 10: 98, worker, soldier (Type locality: Batavia, 

Java). New Synonymy (fide W. L. Brown). 
Pheidole oceanica subsp. cavannae Emery, 1877, Ann. Mus. Stor. Nat. Genova 25: 464, nota, 

soldier (Type locality: New Caledonia) ; 1914b: distribution. New Synonymy. 
Pheidole (Pheidole) oceanica var. nigriscapa Santschi, 1938, Insects of Samoa 5 : 48-49, fig. 

3a, worker, soldier (Original localities : Apia and Pago Pago, Samoa). New Synonymy. 
Pheidole oceanica nigriscapa var. tahitiana Santschi, In Cheesman & Crawley, 1928, Ann. Mag. 

Nat. Hist. ser. 10, 2: 516-517, worker, soldier (Type locality: Tahiti). New Synonymy. 
Pheidole megacephala, auct. (partim), nee Formica megacephala Fabricius, 1793. 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, beating dead branches (ECZ, OHS, 
1940). Tutuila: Pago Pago (A. F. Judd, 1926). TONGA: Tongatabu (W. Cottrell-Dormer). 
Vavau: Neiafu and Holonga (NK, 1956). Eua : Ohonua (NK, 1956). TOKELAU IS.: 
Fakaofo (EHB, 1924). COOK IS. : Rarotonga (G. P. Wilder 1925 ; unknown collector 1937). 
SOCIETY: Tahiti: W. M. Mann (1914). Fautaua Vail., 3.2km, from sea (A. M. Adamson 
1928). Afaahiti, near Vaitaare (ECZ, 1934). Tiupi Bay, Papeari, "household nuisance" 
(ECZ, 1934). 3.2km S of Papeete on coast (RWT, 1960). Moorea: (McTavish ; G. P. 
Wilder 1925 ; H. Smith). *Raiatea (1926). *Huahine (1926). Makatea (G. P. Wilder 1932)! 
Wheeler (1936) records fervens from numerous localities on Tahiti. Cheesman & Crawley 


Fig. 33. Pheidole fervens Fr. Smith. Afiamalu, Upolu (ECZ) : soldier, left and 
upper right; minor worker, lower right. 

Pac. Ins. Mon. 

record it from several of the islands listed, as well as Motu Moute (near Bora Bora), and 
note that it is " extremely abundant on the inhabited shore-belts, where workers swarm in 
the houses and native huts after food." AUSTRAL IS.: Rurutu: 15m, 300m from sea 
(A. M. Stokes 1921). Upopepe Vail., 30m (ECZ, 1934). "Rimatara: Oromana Hills, 80 
m (1934). Raivavae: Mt. Anatoni (1934). Near Unurau, 30m (1934). *GAMBIER IS.: 
Timoe, northern islet (1934). * MARQUES AS : Mohotane, 200 m. UaHuka: Hanatekeo, 
300 m. Hiva Oa. Cheesman & Crawley (1928) also record the species from Fatu Hiva. 
*PITCAIRN: Back of village and S. side of island (1934). 

This species is widespread through tropical Asia, from China to the Moluccas. It is 
spottily distributed in the Pacific Region, where it has apparently been spread by human 
commerce. We have seen samples from Micronesia, the Fijis, and New Caledonia, as well 
as Polynesia. In the past fervens has been consistently confused with oceanica, so that 
published records are frequently mixed. It can be distinguished from oceanica by its smaller 
size and lighter sculpturing. In oceanica the broad area between the eyes and posterior 
limits of the frontal carinae is completely covered with heavy longitudinal rugae ; in fervens 
the same area is covered partly by a feebler, irregular rugoreticulum and (in most series) 
partly by coarse shagreening. Actually, fervens is not close to oceanica at all. In some 
features it more closely resembles megacephala. From the latter species it can be separated 
by the completely rugose head of the soldier, contrasting with the smooth and shining 
condition of the entire occipital region of the megacephala soldier. 

Pheidole (P.) megacephala (Fabricius) Fig. 34. 

Formica megacephala Fabricius, 1793, Ent. Syst. 2: 361, soldier. 

DISTRIBUTION: ONTONG JAVA: Leuaniua (J. G. Bradley 1953). SAMOA: Savai'i: 
Patamea, Puapua, Sili (all NK, 1955). Vaisala, coastal native village, strays on banana- 
case wood imported 2 days previously from Apia (Upolu), 21.111.1962. Upolu: Afiamalu, 
700 m (OHS, 1940) ; berlesates of tree moss at 8 m, ground moss, disturbed rain forest, 
III^IV.1962 (RWT aces. 302, 580 ; GE aces. 45, 48) ; foragers in house (RWT ace. 299) ; 
numerous colonies collected and observed under stones, rotting logs, and in moss or leaf 
litter, disturbed rain forest (RWT aces. 225, 226, 272, 279, 280) ; sweepings from grass and 
herbage at roadside (RWT aces. 230, 263). All RWT records were made during III. 1962. 
Falepuna, berlesate, leaf litter, edge of swamp (TEW, 1956). Fasitooufa, strays on roadside 
weeds in disturbed rain forest ; nest under bark, rotting log, coconut plantation, 29.111.1962 
(RWT aces. 604, 605). Fusi (NK, 1955). Malololelei (NK, 1955, **1924). Sliding Rock 
track, colonies collected under stones and in disused termite carton nests on trees, over- 
grown plantation, 13.111.1962 (RWT aces. 291-296). Tapuellele (NK, 1955). Lauli'i, coastal 
Futu grove (TEW, 1956). Le Mafa, 400m, berlesate of leaf mold, rain forest, 13.VI.1962 
(GE ace. 63). Utumapu, rain forest, colonies in rotting wood, 17.111.1962 (RWT aces. 573, 
577). Vaipoto, rain forest (TEW, 1955). Tutuila : Apia (ECZ, 1940; J. L. Gressitt 1952; 
**1924). Blunt's Point, sweeping Bermuda grass (OHS, 1940). Fagatogo, 300m (OHS, 
1940). Le Mafa, 250m, berlesates from epiphytes and fern roots, rain forest, 12. VII. 1962 
(GE aces. 73, 74, 76). Moloata, 300m, beating dead branches (ECZ, 1940) ; 200m, sweep- 
ing (OHS, 1940). Pago Pago, on Ficus at seashore (OHS, 1940); (A. F. Judd 1924; **1925). 
**Manono: Mulifanua (1925). Listed from Pago Pago as P. teneriffana Forel by Santschi 
(1919). ROTUMA: " Melisa " (H. St. John 1938). DANGER IS.: Motu Kotawa (EHB 

Wilson & Taylor : Ants of Polynesia 

Fig. 34. Pheidole megacephala (Fabricius). Vaipoto, Upolu (ECZ) : soldier 
upper; minor worker, lower. 

1924). Nassau (EHB 1924). LINE IS.: Fanning (S. C. Ball. 1922). Palmyra : E. J. Ford 
(1959) records this species on sprouted coconuts brought from Palmyra to Honolulu in 
1957. SWAINS I. : (ECZ, 1940). COOK IS.: Rarotonga: Aoarua (Wheeler 1914). SOCIETY 
IS. : Raiatea (H. E. Crampton 1908). *AUSTRAL IS. : Tubuai: Mt. Tavaetu, 250m (1934) 
Murivani (1934). SW side of Mt. Taita, 300m (1934). Tapapatauai Islet (1934). 'GAM- 
BIER IS. : Mangareva: NW slope of Mt. Duff, 60-90 m (1934). Pass W of Rikitea (1934). 
Agakauitai (1934). Aukena, NW side (1934). Koravao (1934). Taravai NW side (1934) 
Akamaru, N. side (1934). MARQUESAS : Wheeler (1932b) records the species from numer- 
ous localities on the following islands : Hiva Oa, Tahuata, Fatu Hiva, Mohotane, Nuku 
Hiva, Ua Huka, Ua Pu, Eiao. Cheesman & Crawley (1928) record it as abundant in in- 
habited areas on Nuku Hiva, Fatu Hiva, and Hiva Oa. It appears to be one of the 
dominant ants throughout the Marquesas. Adamson (in Wheeler 1932b) states that it 
" extends its range up to the highest elevations in the Marquesas, though at 900 m its 
numbers begin to diminish." HAWAII : Megacephala apparently occurs on all of the islands 
including even Laysan (Butler 1961) and Midway (Timberlake 1926) up to an elevation 
of 900 m. It is probably the dominant ant species in the lowlands and has been a serious 
economic pest for many years. It is generally attributed by naturalists as being one of 
the agents most destructive to the native Hawaiian insect fauna. According to Illingworth 
(1917), who provided a full account of megacephala in Hawaii together with a general 
bibliography of the species, the exact date of its introduction is unknown. In 1880 Black- 
burn & Kirby stated it to be one of the commonest ants in the islands. 

[S Pac. Ins. Mon. 14 

P. megacephala apparently is native to Africa, where the populations are maximally 
variable. It has been spread by commerce to almost all of the more humid parts of the 

Nuptial nights of megacephala were reported by members of the Hawaiian Entomological 
Society as occurring widely on 14.XII.1934, at various localities on Oahu (Proc. Hawn. 
Ent. Soc. 9: 3). 

Pheidole (P.) oceanica Mayr Fig. 35. 

Pheidole oceanica Mayr, 1866, Sitz. Akad. Wiss. Wien (Math.-Nat. Cl.) 53: 510, worker, 

soldier, queen (Type locality: Ovalau). Type series restricted to soldier and queen, 

Mayr (1870). 
Pheidole oceanica var. pattesoni Mann, 1919, Bull. Mus. Comp. Zool., Harv. 63: 317, worker, 

soldier (Type locality : Graciosa Bay, Santa Cruz I) . New Synonymy. 
Pheidole oceanica var. upoluana Santschi, 1928, Insects of Samoa, 5 : 47, fig. 3b, worker, 

soldier, queen (Original localities: Malololelei, Upolu, and Safune, Savai'i). New 

Pheidole oceanica var. boraborensis Wheeler, 1935, Bull. B. P. Bishop Mus. 113 : 15-16, worker, 

soldier (Type locality : Borabora, Society Is.) New Synonymy. 

DISTRIBUTION : SAMOA : Savai'i : Aopo, ca. 170 m, rain forest, colonies in rotting 
logs (RWT aces. 504-506). Asau, disturbed lowland rain forest, strays and colonies from 
rotting logs (RWT aces. 487, 492, 500, 505). Falealupo, lowland rain forest, strays and 
colony from rotting logs (RWT aces. 426, 441, 443). Gagaifoumauga, rain forest, berlesate 
of moss from logs (GE ace. T4). Lotogo, 350m, rain forest, strays from low shrubbery 
and from branches of Ficus tree, 35 m above ground, nests with larvae and pupae from 
rotting logs and a stump (RWT aces. 453, 458, 472, 475, 477, 485). Ga'utavai, native vil- 
lage, strays from coconut palm trunks (RWT ace. 383). Matega, 65 m, banana clearing, 
edge of rain forest, strays from ground, rotting logs and low vegetation, nests with late 
larvae and pupae in rotting logs, 1 with alate queens, 1 with $$, 20.111.1962 (RWT aces. 
392, 394, 398, 401, 402, 404, 410). Mt. Matafa, 700m, rain forest, strays from foliage in 

Fig. 35. Pheidole oceanica Mayr. Syntypes of " var. boraborensis Wheeler ' 
soldier, left and upper right; minor worker, lower right. 

Wilson & Taylor : Ants of Polynesia 

forest, and from roadside grass (RWT aces. 531, 532, 535). Mt. Matafa Road, 200m, rain 
forest, strays from foliage and under bark on rotting logs, nests with larvae and pupae 
in and under rotting logs and in soil under stone (RWT aces. 361, 365-367, 371, 372, 374). 
Papa, native banana plantation, nest under bark of very dry dadap log (RWT ace. 377). 
Patamea (NK, 1955). Puapua (NK, 1955). **Safune, 300-700 m, low forest (1924). Sama- 
laeulu, coastal rain forest, strays from foliage (RWT ace. 513). Toga, coastal rain forest, 
nest under trunk epiphyte 1m above ground (RWT ace. 380). Upolu : Apia, sweeping 
(ECZ, 1940). **Malololelei, 600m (1924). Saagafou, rain forest, nest under stagfern epi- 
phyte and strays under moss on trees (RWT aces. 610, 611). Satalo, Taro patch, edge of 
rain forest, strays under bark on rotting log (RWT ace. 614). Sinaele, 350m, beating 
(ECZ, 1940). Tanumalala, 350m, rain forest, strays from foliage and nest from rotting 
log (RWT ace. 606, 607). Tapatapao, 300m, under dead bark, winged queen, 21.VIII.1940 
(ECZ). Tiavi, 600m (OHS, ECZ, 1940). Togitogiga, 400m, rain forest berlesates of leaf 
litter and moss from rotting logs (GE aces. 56, 58, 59). Tutuila: Fagatoga, 300m, beating 
dead branches and shrubbery (ECZ, 1940). Moloata, 220m, sweeping dealate queen (ECZ, 
1940). Reservoir at Fangatanga, beating shrubbery, dealate queen (ECZ, 1938). Pago Pago 
(A. F. Judd **1923). Manua: **Ta'u (1923). ELLICE IS.: Mayr (1870). TONGA: 
Tongatabu, record by Mayr (1870). ROTUMA : " Kilinga ", 60m, workers, winged queens, 
4.VIII.1938 (H. St. John). COOK IS.: Rarotonga (G. P. Wilder 1925). TOKELAU IS.: 
Fakaofo (EHB, 1924). DANGER IS.: Motu Kotawa (EHB, 1924). LINE IS.: Fanning (S. 
C. Ball, 1922). SOCIETY IS.: Bora Bora (boraborensis types, H. E. Crampton 1908). Tahiti: 
Papenoo Valley (NK, 1955). Moorea: Baie de Cook (NK, 1955). Raiatea: Uturoa (NK, 
1955). MARQUESAS: Mohotane (Le Bronnec & Tauraa 1931). Ua Huka: Hanatekeo, 
300m, Hane Valley (Le Bronnec & Tauraa 1931). 

The true oceanica (see discussion under P. fervens) ranges continuously from New Guinea 
and New Caledonia northward into Micronesia and eastward to the Marquesas Is. The 
distributional and taxonomic evidence suggests that it is native to most or all of this range. 
Unlike the tramp species megacephala, it does not extend beyond the Marquesas and is 
not known outside the Pacific area. Further, it is closely related to P. impressiceps Mayr, 
a species native to northern Queensland and western Melanesia. 

Previous authors have indicated a considerable amount of geographic variation in the 
soldier caste of oceanica. The characters cited have been total size, body form (especial- 
ly head shape), scape length, form and intensity of sculpturing (especially on the head) 
and color. In the present study extensive series from over the entire range of the species 
have been examined. Variation has been found to be unusually pronounced, but most of 
it is nongeographic. Nearly the entire range of variation in most of the characters can 
be found in several series from Upolu, for example, and large segments of the variational 
range of single characters can be found even in single nest series. Only one character, 
thickness of propodeal spine, seems to show significant geographic differentiation. Soldiers 
in 8 of 9 series examined from Samoa have thicker propodeal spines than members of the 
same caste from all 20 Melanesian series examined, originating from New Guinea, Solomon 
Is., Santa Cruz Is., New Hebrides, and Fiji. Single series from Rotuma, Danger Is., Bora 
Bora, and Marquesas Is. conform to the Samoan type. Thus in the limited material avail- 
able Polynesian series generally can be separated from Melanesian series on the basis of 
this one character. It may be noted that in several Micronesian series thus far examined, 
originating from several parts of the Caroline Is., all soldiers possess the thinner propodeal 

50 Pac. Ins. Mon. 14 

spines characteristic of the Melanesian populations. 

Pheidole (P.) umbonata Mayr Fig. 36. 

Pheidole umbonata Mayr, 1870, Verb. Zool.-bot. Ges. Wien, 20: 978-9, soldier (Type locality: 

Tonga Is.) 
Pheidole (Pheidole} umbonata subsp. zimmermani Wheeler, 1936, B. P. Bishop Mus., Occ. Pap. 

12 : 6-7, soldier, worker, queen, $ (Type locality : near Vaitaare, Papeari, Tahiti). New 


DISTRIBUTION: SAMOA: Savai'i: Alagaogao, ca. 100m, banana plantation, colony 
in rotting log (RWT ace. 418). Asau, lowland rain forest, colonies in rotting logs and 
a stump (RWT aces. 488-9, 501). Falealupo, dry rain forest, strays under bark, dry rott- 
ing log (RWT ace. 436). Lotogo, 350m, rain forest, colony in rotting stump (RWT ace. 
486). Manase, strays ex clay bank (RWT ace. 507). Matega, 65m, banana clearing, edge 
of rain forest, colonies in rotting logs (RWT aces. 395-6, 407). Mt. Matafa Rd., ca. 200 
m, edge of rain forest, strays ex foliage and under bark, rotting log (RWT aces. 361, 365). 
Palauli (NK, 1955). Patamea (NK, 1955). Samalaeula, coastal rain forest, colonies under 
small epiphytes on tree trunks, and in rotting logs (RWT aces. 514, 517-19). Sili (NK, 
1955). Taga, coastal rain forest, ground strays (RWT ace. 381). Upolu: Afiamalu, 700m 
OHS, ECZ, 1940) ; (TEW, 1956) ; undisturbed, disturbed, and partly cleared rain forest, 
strays and colonies from small rotting wood fragments, in and under rotting logs, under 
moss and epiphytes on logs, stumps and tree trunks, and 1 colony each under a stone and 
under moss on ground (RWT aces. 241, 264-6, 268-70, 278, 309, 331-2, 336, 348-9, 601) ; 
disturbed or partly cleared rain forest, berlesates, moss on logs (RWT aces. 2319-23 ; GE 
aces. 46, 53-4); berlesates, tree moss, 8-10 m (RWT ace. 2307; GE aces. 44-5). **Apia 
(1925). Fagaloa, 170m, rain forest, berlesate, epiphytes and litter on rotting log (GE ace. 
68). Falepuna, edge of swamp (TEW, 1956). Le Mafa, 400m, rain forest, strays on road- 
side clay bank and under moss on tree trunk (RWT ace. 634-5) ; berlesates, leaf litter, 

Fig. 36. Pheidole umbonata Mayr. Afiamalu, Upolu (ECZ) : soldier, left and 
upper right ; minor worker, lower right. 

Wilson & Taylor: Ants of Polynesia 

moss, epiphytes (GE aces. 63, 65-6) ; (NK, 1955). Malololelei Road, 550m, dead tree fern 
fronds (OHS, 1940; "1924). Poutasi (TEW, 1956). Sa'agafou, lowland rain forest, colo- 
nies under epiphyte ferns on tree trunks (RWT aces. 609-11). Sinaele, 450m (ECZ, 1940). 
Above Selani Falls, rain forest (TEW, 1956). Tanumulala (TEW, 1956) ; 350m, disturbed 
weed-overgrown rain forest, colonies under moss on tree trunk and in rotting stump (RWT 
aces. 606-7). Tapatapao, 250 m, beating dead branches (ECZ, 1940). Tiapapata, rain forest, 
berlesates, moss on trees and vines (GE aces. 69-71, 73-4). Togitogiga, 400m, rainforest, 
berlesates, debris on rotting log and leaf litter (GE aces. 57, 59). **Tuaefu (1924). **Vai- 
lima (1925). Utumapu (TEW, 1956). Tutuila: North of Aloau, 450m, stunted forest, 
berlesates, moss on logs (GE ace. T16). **Leone Road (1923). Moloata (C. P. Hoyt 1953). 
Mt. Pioa, 550m, beating shrubbery (ECZ). Naval station, winged $$ and queens, 12.VII. 
1940 (OHS, ECZ, 1940). Mayr recorded the species from Samoa in 1876. All Samoan col- 
lections by Taylor (RWT) and Ettershank (GE) were made in March- July 1962. TONGA : 
(Type locality Mayr, 1870). *SOCIETY IS. : Tahiti: near Vaitaare, Papeari. Tiupi Bay, 
Papeari. Moorea: Tehau Point. *AUSTRALIS.: Rurutu: Mt. Manureva, 400m. Raivavae : 
Mt. Hiro, 400 m. *MARQUESAS : Fatu Hiva : Vaikoao, Omoa, 500 m. 

P. umbonata ranges from New Guinea N into the Marianas and E continuously across 
Melanesia to Samoa. It is unknown from central Polynesia but widely dispersed in the 
Society, Austral, and Marquesas Is. Emery (1914) records it from New Caledonia, but 
the record is probably in error, referring to a closely related species that replaces it there. 
The latter species, which is undescribed, can be distinguished from umbonata by its dis- 
tinctly longer antennal scapes, more elongate petiolar peduncle, and several minor sculptural 
characters. On New Guinea umbonata is a relatively scarce species known only from coast- 
al localities. Emery (1900) records it only from Friedrich-Wilhelmshafen (Madang) and 
Beliao I. There is in the Museum of Comparative Zoology a single series from Finsch- 
hafen, collected by N. G. L. Wagner. During a 3-month collecting tour of eastern New 
Guinea, Wilson did not succeed in finding it once. Umbonata is mostly replaced on New 
Guinea by a bewildering array of related endemic species that occur chiefly in the inner 
rain forests and highlands. The densest populations of umbonata are in outer Melanesia 
and Samoa. 

Geographic variation has been noted in two independent characters, worker size and 
coloration. Both show roughly concentric patterns and are notably discordant with respect 
to each other. They can be summarized as follows : 

Size. Soldier and minor workers originating from the central populations (New Britain, 
New Hebrides, Fiji) are slightly smaller than those from the peripheral populations (New 
Guinea, Marianas, Samoa, Tahiti). This pattern is shown in the following soldier head 
width measurements, given in millimeters, for small samples of individuals from each avail- 
able nest series : Finschhafen, New Guinea, 0.80, 0.83, 0.83 ; Ralum, New Britain, 0.76 ; 
Rennell I., Solomons, 0.76, 0.77 ; Sola, Vanua Lava, Banks I., 0.79, 0.79 ; Luganville, New 
Hebrides, 0.79 ; Baldwin Bay, New Hebrides, 0.80, 0.80 ; Lasema, Fiji, 0.76 ; Loma Loma, 
Fiji, 0.76, 0.76, 0.79 ; Somo Somo, Fiji, 0.73 ; Suene, Fiji, 0.77, 0.79, 0.79 ; Agrihan, northern 
Marianas, 0.82, 0.82, 0.83; Afiamalu, Samoa, #1, 0.83,0.84,0.86; Afiamalu, Samoa, #2,0.83, 
0.84, 0.85 ; Malololelei, Samoa, 0.80, 0.82, 0.82, 0.82 ; Sinaele, Samoa, 0.80, 0.82, 0.83 ; Tapa- 
tapao, Samoa, 0.80, 0.84 ; Vaitaare, Tahiti, 0.77, 0.82, 0.86. (Maximum error of measurement 
= 0.01 mm). 

52 Pac. Ins. Mon. 14 

Color. Soldiers and minor workers originating from central areas and from the northern 
periphery (Marianas) tend to be lighter in coloration. Series from Rennell I. (Solomons), 
Banks Is., New Hebrides, Samoa, and the Marianas are very light reddish brown to red- 
dish yellow. The lightest of all are from the Marianas (Agrihan). Occasional Samoan 
nest samples tend to a darker shade, approaching medium reddish brown. Series from 
Ralum, New Britain, and Tahiti, which are more peripherally located, are medium to dark 
reddish brown. A single series from Finschhafen, New Guinea, is an intermediate shade 
that can perhaps best be described as light reddish brown. 

Pheidole (Pheidolacanthinus) sexspinosa Mayr Fig. 37. 

Pheidole sexspinosus Mayr, 1870, Verh. zool.-bot. Ges. Wien, pp. 977-978, worker, soldier 

(Type locality : Ellice Is.) 
Pheidole (Pheidolacanthinus} sexspinosa subsp. adamsoni Wheeler, 1932, Bull. B. P. Bishop 

Mus. 98: 157, worker, soldier, queen, <^ (Type locality: Vaikoao, Omoa Valley, Fatu 

Hiva, Marquesas). New Synonymy. 

DISTRIBUTION : SAMOA : Savai'i : Asau, 170 m, strays on low foliage, rain forest 
(RWT and GE). Falealup, dry rain forest, strays from forest floor and low foliage 

Fig. 37. Pheidole sexspinosa Mayr. Syntypes of "subsp. adamsoni Whee- 
ler" from Fatu Hiva, Marquesas : soldier, upper ; minor worker, lower. 

Wilson & Taylor : Ants of Polynesia 

(RWT aces. 424, 427, 440). Le Mafa, 400m, strays under small trunk epiphytes (GE ace. 
66). Lotoga, 330 m, rain forest, strays from tree trunk (RWT ace. 477). Matega, 65 m, 
banana clearing, forest edge, nest with alates under bark on log (RWT ace. 393) ; strays 
from rain forest floor and rotting stump (RWT aces. 390, 400, 409). Mt. Matafa Rd., 100 
m, rain forest, strays from low foliage and under bark of rotting log (RWT ace. 361, 
365). Mt. Matafa, 350 m, rain forest, nests under small trunk epiphyte and in rotting 
log (RWT ace. 530, 533). Puapua (NK, 1955). Samalaeulu, coastal rain forest, strays 
under trunk epiphytes (RWT ace. 510). All RWT and GE accessions from Savai'i were 
made in III. 1962. Upolu : Afiamalu, 700 m, workers and winged queen, beating shrubs and 
under dead bark, 25.VII.1940 (ECZ) ; colony under log, Taro patch, 16.111.1962 (RWT ace. 
322) ; partly cleared rain forest, ground foragers, 12.111.1962 (RWT ace. 265) ; berlesate, 
disturbed rain forest, moss on logs, 9.IV.1962 (GE ace. 54). Mafa Pass Road (NK, 1955); 
Malololelei, 500m (ECZ, 1924, **1924). Tutuila: West side, Afono Trail, 150m, (EHB 
1935; ECZ 1940). Amouli, 300m, (ECZ, 1940 **1923). **Fagasa (1924). Fagotogo, 
300m, beating shrubbery (ECZ, 1940). **Leone Rd. (1923). " Moloata ", 300m, beating 
dead branches (ECZ 1940). Matafao Trail, 400m, beating shrubbery (ECZ 1940). Mt. Pi- 
oa, 550 m, beating shrubbery (ECZ 1940). N. side, Pago Pago, 400 m, beating dead branches 
(ECZ, 1940). Pago Pago (NK, 1941; **1924). ELLICE IS.: type locality (Mayr 1870). 
SOCIETY IS.: Tahiti: Papeari (A. M. Adamson 1928), adamsoni syntypes. MARQUE- 
SAS: Fatu Hiva: Vaikao, adamsoni syntypes (Le Bronnec 1931). 

Sexspinosa occurs continuously in Melanesia from New Guinea to the New Hebrides 
and northward into the Caroline and Marshall Is. It is absent from Fiji and New Cale- 
donia. It appears to be rather common in Samoa, especially on Tutuila, but is scarce 
in the remainder of Polynesia. 

Geographic variation is shown by at least 2 independent characters in the soldier and 
worker castes. The body color in the 2 Society and Marquesas series is piceous brown. 
In 16 Melanesian and Samoan series it ranges from medium to dark reddish brown. In 
5 Micronesian series it is dark reddish brown, averaging intermediate in shade between 
the Melanesian-Samoan and outer Polynesian samples. The Society-Marquesas sample also 
differs from the other material examined in having slightly thicker propodeal spines, a 
geographic trend paralleling that found in the soldier caste of P. oceanica. Contrary to 
Wheeler's (1932b) statement, no significant geographic variation appears to be shown in 
head form or in the sculpturing of head and thoracic dorsum in the soldier caste. These 
features vary, especially pronotal sculpturing, but the variation is non-geographic and may 
even be in large part intranidal. 

P. sexspinosa belongs to a group of species maximally diversified on New Guinea. The 
nature of geographic variation in the species seems to indicate that it is native through- 
out most orall of its present range in the Pacific. 

Cardiocondyla emeryi Forel 

Fig. 39. 

Cardiocondyla emeryi Forel, 1881, Mitt. Miinchen Ent. Ver. 5: 5, worker. 
Cardiocondyla nuda subsp. nereis Wheeler, 1927, Proc. Amer Acad. Arts Sci. 62 : 140, work- 
er, queen (Type locality : Norfolk I.) New Synonymy. 

DISTRIBUTION : SAMOA : Upolu : Afiamalu, 700 m, several series swept from grass 
and weeds in overgrown European garden, and along roadside in disturbed rain forest 

Pac. Ins. Mon. 

Fig. 38. Cardiocondyla nuda (Mayr). Fig. 39. C. emeryi Forel. Fig. 40. C. wroughtoni (Fo- 
rel). Workers from United States. (After M. R. Smith, 1944.) 

(RWT aces. 230, 263, 287, 297). SOCIETY: Tahiti: Faa, 6km from sea, 300m, sweep- 
ing over grasses (A. M. Adamson 1928). Hitiaa, 5 km from sea, 300 m, from Metrosi- 
deros, winged queen, 20.11.1928 (Adamson). Papenoo Vail. (NK 1955). Moorea: Tehau 
Point, near sea level, sweeping grasses and low herbage (ECZ 1934). Baie de Cook (NK 
1955). Raiatea: Opoa (NK 1955). Uturoa (NK, 1955). Cheesman & Crawley (1928) 
found emeryi in 2 localities near Papeete. Wheeler (1936) records "nuda subsp. nereis", 
presumably this species, from numerous additional localities on Tahiti and Bora Bora. 
AUSTRAL IS. : Tubuai : Murivahi, near sea level, sweeping grasses and low herbage (ECZ, 
1934). Raivavae: near Anatonu, near sea level, sweeping grasses and low herbage (ECZ, 
1934). Near Ahuoivi Point, sea level (ECZ 1934). Rimatara: Anapoto, 8m, sweeping 
grasses and low herbage (ECZ 1934). Wheeler (1936) records numerous additional series 
collected by Zimmerman in 1934 on most of the principal Austral Is. : Rimatara, Maria 
I., Raivavae, Tubuai, Rurutu, Morotiri. It was clearly a dominant species during at least 
the time of Zimmerman's visit. *TUOMOTU : Tukuhora: Anaa (ECZ, 1934). GAM- 
BIER IS. : Mangareva: Point Teone Kura, 400m (ECZ, 1934). Agakauitai: sea level 
(ECZ, 1934). Aukena: Korovao (ECZ, 1934). Taravai : winged queen, 1.VI.1934 (ECZ). 
All of these samples were taken by sweeping grasses and low herbage. MARQUESAS : 
Eiao: 500 m (Le Bronnec & Tauraa 1931). Hiva Oa: Kopaafaa, 900m (Mumford & Adam- 
son 1927). Tanaeka Val., 450 m, dry slope (Mumford & Adamson] 1929). Ua Huka: 
Tauheeputa, 550 m (Le Bronnec & Tauraa 1931). Wheeler (1932b) records this species 
from numerous additional samples taken during 1927-1931 by Adamson, Le Bronnec, Mum- 
ford, and Tauraa in the following islands : Hiva Oa, Tahuata, Fatu Hiva, Mohotane, Nuku 
Hiva, Ua Huka, Eiao, and Hatutaa. Emeryi was clearly one of the most abundant ants 
throughout the Marquesas at least during this time. PITCAIRN : N. side (ECZ 1934). 
*HENDERSON: N and NW side (ECZ 1934). OENO : (ECZ 1934). HAWAII : Swezey 

1967 Wilson & Taylor : Ants of Polynesia 55 

(1943) reports that M. R. Smith determined a majority of the series of Cardiocondyla col- 
lected in the Hawaiian Is. as emeryi. No particular islands were cited. As noted by 
Swezey, emeryi was not recorded by Wheeler (1934a) in abundant collections of Cardio- 
condyla made up to the early thirties. Since emeryi is a distinctive species, the chances 
are great that it first gained a secure foothold in the islands during the thirties. 

C. emeryi is a native African species that has been distributed by commerce throughout 
the tropics. It is closely related to the better known C. nuda (Mayr) and has been con- 
sistently confused with that species by past authors. In the Pacific Region at least, emeryi 
can be distinguished from nuda by the following worker characters: (1) the thoracic dor- 
sum is relatively finely and sparsely punctate with its surface feebly shining, as opposed 
to densely punctate and opaque in nuda ; (2) the humeri are slightly more produced and 
angulate ; (3) the petiolar node is distinctly more slender and from directly above ap- 
pears longitudinally elliptical, as opposed to subcircular in nuda. Also, the overall pigmen- 
tation averages lighter and the propodeal spines somewhat longer in emeryi, although the 
ranges of these characters overlap in the 2 species. 

Cardiocondyla nuda (Mayr) Fig. 38. 

Leptothorax nudus Mayr, 1866, Sitz. Akad. Wiss. Wien, Math. Nat. 53: 508, worker (Type 
locality: Ovalau, Fiji). M. R. Smith, 1944, Proc. Ent. Soc. Wash. 46: 38, taxonomy, 
Cardiocondyla nuda var. minutior Forel, 1899, Fauna Hawaiiensis (Heterogyna), p. 120, 

worker (Original localities: Hawaii and Malokai). New Synonymy. 
DISTRIBUTION: SAMOA : Savai'i: Fanga (NK, 1955). Palauli (NK, 1955). Patamea 
(NK, 1955). Puapua (NK, 1955). **Safune (1924). Upolu : Afiamalu, 700m, under dead 
bark and sweeping Commelina, several series (ECZ, 1940). Alagaogao, 100m, strays from 
roadside weeds (RWT ace. 416). Falefa (NK, 1955). Lafanga (NK, 1955). Le Mafa, 
nest from soil of clay roadside cutting undisturbed rain forest (RWT ace. 635). Tuiuila: 
Aua, seashore, sweeping Sporobolm (OHS, 1940). Fagatogo, 300m (OHS, 1940). Pago 
Pago (W. M. Wheeler 1931). Vailoa (D. T. Fullaway 1930). **Manua: Ta'u (1923). Mayr 
recorded nuda from the islands in 1876. TONGA: Tongatabu: Nukualofa (NK, 1956). 
Mayr (1876). COOK IS.: Rarotonga (G. P. Wilder, 1925). DANGER IS.: Motu Kotawa 
(EHB, 1924). SOCIETY IS. : Raiatea : Opoa (NK, 1955). AUSTRAL IS. : Rimatara: 
Oromana Hills, 80m, sweeping grasses and low herbage (ECZ 1934). *MARQUESAS : 
Fatu Hiva : Omoa, Wheeler (1932b) cites a worker "clearly belonging to the typical form 
of the species." LINE ISLANDS : What may be this species was recorded from Johnston 
I. by M. R. Smith as "Cardiocondyla sp." (in Chilson 1953). *FLINT I.: Wheeler (1934b) 
recorded "var. minutior." EASTER I.: Wheeler (1922) recorded the species as "var. minu- 
tior." I. E. Efford and J. A. Mathias collected it at the following localities in January 
1965 : on grass on hillside at Ovahe, winged queen ; "on vegetation in Cabin" ; Rano Ra- 
raku, winged queens. HAWAII: Oahu: Honolulu (1910, no collector given) ; no locality 
(van Zwaluwenburg, probably 1925-33). Hawaii: Kilauea, 1300m. (Wheeler, 1930). Forel 
(1899) records the species (as "var. minutior") from Hawaii and Molokai. Records from 
Midway, Necker, and French Frigate Shoals are given by Timberlake (1926); the collections 
were made in 1923 by Bryan & Fullaway. Between this date and the early 1930's nu- 
merous other collections (recorded in Wheeler 1934a) were made on Oahu, Molokai, La- 

56 Pac. Ins. Mon. 14 

nai, Maui, Kauai, Hawaii, Midway, Necker I., and French Frigates Shoal. Bryan (1935) 
records it from little (600m diameter) Rabbit I. near Oahu, and Butler (1961) mentions 
specimens collected on Laysan I. in 1959. Butler also collected the species on Kure I. in 
September 1961. 

C. nuda is a species of probably African origin that has been spread by commerce 
throughout the warmer parts of the world. 

Cardiocondyla wroughtoni (Forel) Fig. 40. 

Emeryia wroughtonii Forel, 1890, Ann. Ent. Belg. (Comptes-Rend. Seances) 34: 110, erga- 
tomorphic $ (described as worker) (Type locality: Poona, India). 

Cardiocondyla wroughtonii var. hawaiensis [!] Forel, 1899, Fauna Hawaiiensis (Formicidae), 

p. 119, worker (Type locality: Molokai). New Synonymy. 
DISTRIBUTION: HAWAII: Oahu (R. H. van Zwaluwenburg). Several specimens from 

flower spikes and seed pods collected 1940-1948 (Ex USNM collection). Between the 

years 1899 and 1934 wroughtoni was collected on the following islands (records in Wheeler 

1934a) : Molokai, Oahu, Kauai, Maui, and Hawaii. 

Wroughtoni is evidently native to tropical Asia. A single collection has been made on 
the beach at Ralum, New Britain (Forel 1901), but the species does not seem to be native 
E of Formosa and the Asian mainland. Instead, it is replaced by the related species ni- 
grocerea Karawajew in the Moluccas and nivalis Mann in the Solomon Is. Wroughtoni oc- 
curs probably as an introduced population in the Carolines and Marianas. It has also 
been spread by commerce to southern Florida in the United States. 

Vollenhovia pacifica Wilson and Taylor, new species 

Fig. 41. 

DIAGNOSIS : A Samoan population closely resembling the widespread Indo-Australian 
oblonga (Fr. Smith) (=oblonga pedestris auct.} which ranges from tropical Asia to New 
Hebrides and New Caledonia. Pacifica differs from oblonga chiefly in its smaller average 
size, proportionately longer head, and the following feature of cephalic sculpturing : lon- 
gitudinal striae limited to median 1/4 of head, their interspaces containing punctures that 
do not distort striae (in oblonga striae covering almost entire anterior surface and in many 
places curve around large, conspicuous interstitial foveae). 

Holotype worker: HW 0.61 mm, HL 0.74 mm, CI 82, SL 0.47 mm, SI 77, EL 0.15 mm, 

PW 0.50 mm. Body form as shown in 

i,, | | \\ j 1\ / . fig. 41. Except for head proportions these 

W^^^^ /^ figures fit most specimens of oblonga also. 

( \ \ \ / ^^I^U-^yL But note subpetiolar process ; it is sub- 

^_^_ / \ ^-/Y /c /S an g u l ar an d with a straight posterior 

'XT^T "^x ^^----^C-U -A-'-?---./ border ; in most oblonga it is smoothly 
, . f TT^^ rounded behind and with a slightly convex 

posterior border. Head of Pacifica holo- 

Fig. 41. Vollenhovia pacifica Wilson and type with median longitudinal striae extend - 

Taylor, holotype worker. i ng to occiput, diverging prominently from 

midline near frontal carinae and near occiput. Strial patch is about 0.22 mm across, or 

the distance between frontal carinae. Interstrial spaces less than 0.01 mm across and filled 

Wilson & Taylor : Ants of Polynesia 

with irregular punctures and shagreening. Away from striae, sides of head are densely 
covered with nearly perfectly round, flat-bottomed foveae 0.01-0.02 mm in diameter; 
no particular orientation of intervening rugulae noticeable. Pronotum very feebly and 
sparsely foveolate and rugulate, feebly shining; in contrast with remainder of alitrunk, 
which is heavily sculptured and opaque. In other respects resembling a small oblonga 
worker. As in oblonga, the body is concolorous blackish brown and the appendages con- 
trasting brownish yellow. 

Paratypes : Worker : HW 0.56-0.60 mm. The series is very uniform in almost all of 
the characters cited with reference to the holotype. The subpetiolar process varies from 
somewhat more angulate than the holotype to as blunted and rounded as the extreme end 
of variation in oblonga in the direction of pacifica. 

Queen : HW 0.72 mm, HL 0.82 mm, CI 88, SL 0.55 mm, SI 76. The queen of the holo- 
type nest series. A normal, formerly alate individual sharing diagnostic characters of the 
species cited for worker. In addition to the usual caste difference, the queen differs in a 
more prominent development of rugulae on the alitrunk and pedicel ; these form a con- 
spicuous concentric pattern around the propodeal spiracle. 

Q-. HW (across and including eye) 0.58 mm. Distinguished from 8*$ of sympatric 
samoensis by its jet black body coloration (medium yellowish brown in samoensis} and 
densely sculptured, opaque head and pronotum (these parts mostly smooth and shining in 
samoensis) . 

MATERIAL EXAMINED : SAMOA : Upolu : Afiamalu, 700 m, holotype worker, 1 1 workers, 
queen, 2&&, 9.III.1962 (RWT ace. 255) ; 6 workers, 9.III.1962 (RWT ace. 236) ; 15 work- 
ers, 10.111.1962 (RWT ace. 244) ; 4 workers, 10.111.1962 (RWT ace. 245) ; 15 workers, 30. 
III. 1962 (RWT ace. 632) ; berlesates, moss on logs, disturbed rain forest III-IV 1962 (RWT 
aces. 2319, 2320 ; GE ace. 47) ; berlesates, tree moss, 8-9 km, disturbed rain forest, III-IV 
1962 (RWT ace. 2312; GE ace. 45). 

Pacifica seems best regarded as being a peripheral cognate of oblonga. It is interesting 
that specimens of oblonga from New Caledonia, which is geographically relatively close 
to Samoa, are nevertheless more similar to the "typical" oblonga from central Melanesian 

Vollenhovia samoensis Mayr Fig. 42. 

Vollenhovia samoensis Mayr, 1876, J. Mus. Godeffroy, Hamburg 12 : 44-45, worker, ft. 

(Type locality: Samoa). 

Vollenhovia agilis Santschi, 1928, Insects of Samoa 5 (1) : 44-45, fig. 1, queen (Type local- 
ity : Fagasa, Tutuila, Samoa). New Synonymy. 

DISTRIBUTION : SAMOA : Savai'i : Gagaifoumauga, rain forest, berlesate, frass under 
bark of newly fallen tree, 8 m above ground, 17.V.1962 (GE ace. SI). Matega, 600 m, 
edge of rain forest, under moss on rotting log, 20.111.1962 (RWT). Upolu : Afiamalu, 700 
m, beating shrubbery (ECZ, 1940) ; disturbed rain forest, nest in moss on tree 2 m from 
ground (RWT ace. 235) ; same as ace. 235, strays in berlesate ; disturbed rain forest, nest 
in rotting log on ground (RWT ace. 298) ; rain forest, wet leaf mold in leaf bases of 
bromeliad low on tree (RWT ace. 306) ; Taro patch in forested area, nest in rotting tree 
fern (RWT ace. 318) ; rain forest, under moss in galleries of wire worms (RWT ace. 

Pac. Ins. Mon. 

Fig. 42. Vollenhovia samoensis Mayr, worker from Afiamalu, Upolu (ECZ). 

602) ; disturbed rain forest, under moss on fallen log (RWT ace. 632) ; all Afiamalu re- 
cords by R. W. Taylor made in III-IV.1962. Malololelei Road, 500m (OHS, 1940; NK, 
1955). Togitogiga, 400m, rain forest, berlesate, moss on rotting log, 12.V.1962 (GE ace. 
60). Tutuila: N of Aloau, 450m, stunted forest, berlesate, moss on logs (GE ace. T19). 
**Fagasa (1924). Matafao Trail, 400 m, beating dead branches (ECZ, 1940). Naval Sta- 
tion, $$ at light, 19.VII.1940 (OHS). Between Oloaua and Olomanu, rain forest, berlesate 
of moss and litter from rotting log, 13.VI.1962 (GE ace. T13). 

V. agilis Santschi is clearly the queen of V. samoensis. These two forms, corresponding 
exactly to the two castes, have been found associated in our Afiamalu series. 

Samoensis retains its status as one of the few ant species truly endemic to Samoa. Even 
so, it is very close to denticulata Emery of New Caledonia, New Hebrides, and Fiji. (We 
have recently examined series of denticulata from Lasema, Fiji, 1915-16, collected by W. 
M. Mann; and Santo, Espiritu Santo, New Hebrides, 22.VIII.1958, collected by Borys 
Malkin. These are the first records from outside New Caledonia.) Samoensis appears to 
differ from denticulata only in its distinctly lighter body sculpturing. In particular the 
pronotum is sparsely punctate and feebly shining, as opposed to longitudinally striate, heav- 
ily shagreened, and subopaque in denticulata. The latter characters are maximal in the 
New Caledonian and New Hebridian denticulata. Statistical studies on larger samples in 
the future may force the synonymy of denticulata with samoensis. 

These 2 species in turn more distantly resemble brachycera Emery of New Guinea, dif- 
fering in their lighter coloration (light reddish brown opposed to piceous brown), distinct- 
ly broader subpetiolar processes, and somewhat longer antennal scapes. They are also re- 
lated to dentata Mann of the Solomon Is., which is distinguished by its slightly larger size, 
denser and coarser mesosomal sculpturing, and total lack of propodeal teeth. Together, 
the 4 forms present the classical pattern of an insular superspecies, with the most far- 
flung pair, denticulata and samoensis, being the least differentiated between themselves. 

At Afiamalu Taylor found samoensis to be relatively abundant in rain forest in various 
degrees of disturbance by man. The species nested on the ground, in rotting logs and 
under moss, and in tree trunks under moss and in epiphytes. His impression was that it 
favored the arboreal sites more than V. pacifica at the same locality. 

Solenopsis geminata (Fabricius) 

Fig. 43. 

Atta geminata Fabricius, 1804, Syst. Piez., p. 423, worker. 
Ana rufa Jerdon, 1851, Madras J. Lit. Sci. 17: 106, worker. 

Wilson & Taylor : Ants of Polynesia 

DISTRIBUTION: TONGA: Tongatabu: Nukualofa, workers, &, 11.1956 (NK) ; W. 
Cottrell-Darmer, no date, probably sometime before 1935. Vavau: Falevai (NK 1956). Eua : 
Ohonua (NK 1956). COOK IS.: Rarotonga: Avarua (W. M. Wheeler 1914). No further 
locality (no collector given, 1924; G. P. Wilder 1925; C. E. Clarke 1937). SOCIETY: 
Tahiti: (record in Mayr 1876). 1.6km S of Papeete, under stones by road (RWT, 1960). 
*Vallee de la Reine (1929). Hitiaa, 300 m (1929). *Fautaua Vail. (1929, 1934). *Arihiri, 
Pare (1934). *Tiupi Bay, Papeari (1934). *Arne. Moorea: Baie de Cook (NK, 1955). 
*Faaroa Vail., 300m (1934). *Tehau Pt. (1934). No locality, (1906-07). Raiatea: Utoroa 
(NK, 1955). *Paaoi Vail., 250m (1934). Cheesman & Crawley (1928): "[In 1925] Ta- 
hiti and Raiatea, abundant on shore-belts on Tahiti, inland on the coast-hill behind Pape- 
ete at 1500 ft." Taylor found the species to be extremely abundant in the environs of Pa- 
peete in 1960. *AUSTRAL IS.: Rapa: Mont Tanga, 250 m (1934). *FLINT I.: (1934). 
TUAMOTU IS. : Fakarava : "Abundant on the beach feeding upon dead molluscs, II-2-25" 
(Cheesman & Crawley 1928). 

LINE IS. :Johnston(,\947-52~),fide , . ,, , ,1. { , ,, 

Chilson (1953). HAWAII: A- ^^W^^jLuJ,, M' 

bundant on Oahu (*1934a ; Ito, \j/f fi]s^_\ \^ * * 
1943), especially in pineapple ^jjl W^'^~^~ / ^~-^J /^ 

fields, sugarcane fields, and dry n/T~~SW^ \\ ^N 

coastal regions (Ito). Carter MT x\ ~* -i ,v\ 'v^ 

(1957) reported the species as 
having appeared for the first A f ig ' 43 D Solenopsis geminata (Fabr] ^ minor woiker from 

& ^ . Aturoa, Raiatea, Society Islands (NK). 

time on Molokai (Naiwa-Apana 

area) around 1956. Bryan (1935) records it from Little Rabbit I., near Oahu, in 1934. 
Suehiro records it for the first time from Midway in 1960. 

S. geminata is native to the tropics and warm North Temperate Zone of the New World. 
A light reddish form ("subsp. rufa") has been spread by commerce throughout the moister 
parts of the Indo-Australian tropics and the darker " typical " form has established itself 
in Africa. In the New World the rufa variant occurs, and produces continuous intergrades 
with the dark variant, from the southern United States to British Honduras and Guate- 
mala (Creighton 1930). Rufa has been characterized by past authors as possessing a me- 
sosternal spine or tooth, but in the material we have examined the structure lacks con- 
stancy. In either case the uniformity in coloration of the respective Indo-Australian and 
African samples and the differences between them illustrate nicely the genetic "bottleneck" 
effect in recently founded populations. It is interesting further to note that the rufa va- 
riant of geminata was first described by Jerdon from India in 1851, while it was first re- 
corded from Tahiti (by Mayr) in 1876. Thus geminata has had a respectably long his- 
tory as a tramp species. Since colonies ordinarily mature in 1 or 2 years, it is fair to say 
that in the case of the Indo-Australian population 50 or more generations have not been 
sufficient to produce a significant differentiation in coloration or any other obvious ex- 
ternal character. 

Geminata favors the drier habitats in Polynesia. As Phillips (1934) and subsequent au- 
thors have pointed out, it is usually replaced in moister, more favorable areas by the more 
widespread Pheidole megacephala. 

Pac. Ins. Mon. 

Solenopsis papuana Emery Fig. 44. 

Solenopsis papuana Emery, 1900, Termeszetr. Fuz. 23 : 330, worker, queen (Original locali- 
ties : Berlinhafen (Aitape) and Friedrich-Wilhelmshafen (Madang), NE New Guinea). 
Solenopsis Dahlii Forel, 1901, Mitt. Zool. Mus. Berl. 2: 14-15, worker ^ (Type locality: 

Ralum, New Britain). New Synonymy. 
Solenopsis cleptis Mann, 1919, Bull. Mus. Comp. Zool., Harv. 63: 330-31, worker (Type 

locality: Wai-ai, San Cristoval, Solomon Is.). New Synonymy. 
Solenopsis cleptes var. vitiensis Mann, 1921, Bull. Mus. Comp. Zool., Harv. 64: 444-45, 

worker (Type locality: Nadarivatu, Viti Levu, Fiji Is.). New Synonymy. 
DISTRIBUTION: SAMOA : Savai'i : Asau, 170m, rain forest, nest and strays from 
rotting logs, V.1962 (RWT aces. 502-504). Gagaifoumauga, rain forest, berlesates of leaf 
mold and soil around tree roots, 14.V.1962 (GE). Letui, lowland rain forest, berlesate 
of leaf mold and soil (RWT ace. 597). Mt. Matafa Rd., 200 m, nest from small rotting 
wood fragment on forest floor, and strays from rotting wood (RWT ace. 364, 365). Upolu : 
Afiamalu, 700 m, disturbed rain forest : colonies and strays from rotting wood (small frag- 
ments to large logs), under moss on logs, and berlesates of ground moss, leaf mold, 

and moss from tree trunks 7-10 m 
I' ' \ \ , , from ground ; III-IV. 1962 (RWT 

-^ i ~^T\ accs ' 228 ' 229 ' 262 ' 310> 347> 2317 ' 

~^^ "^^-V 2320 ' 2321> 2323> 2324; GE accs - 

{J^g- 46-48, 50, 51, 53, 54). Le Mafa, 

/ ^^~^^f 400 m, rain forest, berlesates of 

u / 44 V\ N;- leaf litter in rock cleft, 13.VI. 

^A \V /\^~ 1962 ^ GE acc ' 64 )' Ta P ata P ao 

V >ClU^ 300 m, under dead bark (ECZ 
^X 5 1940). Tiapapata, rain forest, ber- 
^ lesates of moss on trees and 

Fig. 44. Solenopsis papuana Emery, worker from Ta- vines 19 VI 1962 (GE accs 69 
patapao, Upolu (ECZ). _. . . _.. 

70, 74, 76). Togitogiga, 300 m, 

rain forest, berlesates of frass at base of orchid and of leaf litter, 12.VI.1962 (GE accs. 55, 
57). Tutuila: between Oloava and Olomanu, berlesate of moss on rotting log, 13.VI.1962 
(GE acc. T12). 

Mann's cleptis and vitiensis, described from the Solomons and Fiji respectively, were 
erroneously compared with maxillosa Emery rather than papuana. We have been unable 
to find any significant differences of species quality between extensive series of papuana 
from New Guinea and "cleptis" from eastern Melanesia. Fijian and Samoan workers 
differ from New Guinea workers in average coloration. They are uniformly light reddish 
yellow, while New Guinea samples range (by nest series) from light reddish yellow to 
medium reddish brown. 

The new Samoan records, made mostly from Berlese funnel collections, show this in- 
conspicuous species to be a prominent member of the cryptozoa of the Samoan rain for- 

Oligomyrmex atomus Emery 

Fig. 45. 

Oligomyrmex atomus Emery, 1900, Term6szetr. Fuz. 23 : 328-29, pi. 8, fig. 30, worker, sol- 

1967 Wilson & Taylor : Ants of Polynesia i 

dier (Original localities (New Guinea) : Hansemann Mts. ; Tamara I.; Beliao). 
DISTRIBUTION: SAMOA : Savai'i : Asau, 170 m, rain forest, in rotting log, (RWT 
and GE ace. 503). Falealupo, lowland rain forest, from fragment of rotting wood on 
forest floor (RWT and GE ace. 439). Samalaeulu, rain forest, colony nesting under 
small epiphyte on tree trunk (RWT and GE ace. 511). Upolu : Fagaloa, 170m, ber- 
lesate, rain forest, moss on rotting log (GE ace. 67). Le Mafa, 400 m, berlesates of 
ground moss and soil (GE aces. 61, 65). Utumapu, rain forest, nest under small epiphyte 
2m from ground (RWT ace. 570). Tiapapata, berlesates, rain forest, moss in trunks of 
living trees (GE ace. 76). Togitogiga, 400 m, berlesates, rain forest, leaf litter (GE aces. 
57, 59). (All Taylor & Ettershank records III-VI. 1962). Lauli'i (TEW 1956). Tanumalala, 
leaf mold in rain forest (TEW 1956). Poutasi (TEW 1956). Tutuila: Alega, leaf litter in 
coastal Futu grove (TEW 1956). 



Fig. 45. Oligomyrmex atomus Emery. Above left and below, minor work- 
er from Utumapu, Upolu (RWT). Above right, soldier from Falealupo, Sa- 
vai'i (RWT). 

Minor workers of O. atomus are among the smallest ants in the world. The species 
was collected in 1962 at Lomolagi, Viti Levu, Fiji, by Taylor, thus closing the gap in its 
known distribution from New Guinea to Samoa. Despite fairly extensive collecting it has 
not yet been found in the New Hebrides. The Fiji-Samoa workers can be distinguished 
from those originating in New Guinea and the Solomons by the following 2 characters : 
pronotum and mesonotum smoother and more shining in Fiji-Samoan samples ; occipital 
horns of soldier shorter (about 0.04 mm long as opposed to 0.05 mm in New Guinea 

Pac. Ins. Mon. 

and Solomons specimens). This difference is perhaps borderline to specific distinctness. 
We prefer the conservative interpretation of considering all the populations conspecific. 
No discernible geographic variation occurs within the Fiji-Samoa samples. 

Oligomyrmex tahitiensis Wheeler 

Fig. 46. 

Oligomyrmex tahitiensis Wheeler, 1936, B. P. Bishop Mus. Occ. Pap. 12: 11-12, queen, $ 

(Type locality: Taopiri, Mt Aorai, "3500-4500 ft.", Tahiti). 

The species is known only from the type collection of the winged queen and 2S& from 
--^MI/iii, , i,,, Tahiti. Unfortunately, most of the many Indo-Australian Oligomyr- 

^ * . * \s mex s P ecies are known only from the worker and soldier castes, 
~f o K which are often radically different from the queen, so that the 

\ fl fe; status of tahitiensis cannot be established at this time. We have 

I //f| been able ' nowever ' to compare the syntype with definitely asso- 

A 1 s\ V/ Y ciated queen of atomus and can ascertain at least that it is not that 
_ R species. In the following description we have given the measure- 
x, ^ , , i , ments of the type and compared it critically with the New Guin- 

^ {^/ 46 ea Q ueens of the wide-ranging atomus, in the hope that generally 
Fig. 46 Oligomyr- useful characters will be highlighted. For further details, in the 
mex tahitiensis Wheel- absence f comparative material, the more complete original des- 
er, syntype queen. cription should be consulted. 

Queen syntype : HW 0.80 mm, HL 0.90 mm, CI 89, SL 0.50 mm, SI 60, EL 0.26 mm. 
(HW of atomus queen is only about 0.42 mm, but the eye is proportionately much larger, 
with EL =0.20 mm). Occipital margin deeply concave, depth at midline (difference be- 
tween transects of occipital corners and occipital midline) about 0.05 mm, in contrast to 
feebly concave border of atomus. In side view the dorsal and basal propodeal faces 
meet in a right angle (evenly rounded in atomus}. Subapical process of petiole better de- 
veloped, consisting of a distinct, acute, anteriorly directed tooth beneath anterior peduncle. 
All of the head evenly, longitudinally striate, except the occiput for a distance of about 
0.2 mm from corners, region posterior to eyes, and gulae, which are very smooth and 
shining. Entire propodeum, metapleuron, petiole, and postpetiole shagreened and opaque : 
remainder of mesosoma and gaster very smooth and shining. Legs and antennae brown- 
ish yellow ; body nearly concolorous rich reddish brown. 

The record of this Oligomyrmex species from Tahiti is an anomaly, since only O. ato- 
mus is otherwise known as far east as Samoa. No judgment can be made on its status 
until the worker caste is obtained and studied with reference to the Indo-Australian spe- 

Chelaner antarcticus (White) 

Fig. 47. 

Formica antarctica White, 1848, Voy. Erebus and Terror, Zool. : Insects, pi. 7, fig. 13, alate 

$ (Type locality: New Zealand). 
Monomorium (Notomyrmex) rapaense Wheeler, 1936, Occ. Papers, B. P. Bishop Mus. 12 : 

10-11, worker (Type locality: NE ridge of Mangaoa Peak, 300-400 m, Rapa). (Syn- 

types examined MCZ). New Synonymy. 
Monomorium antarcticum : Brown, 1958, Acta Hymenopt. 1 : 29, fig. 4, taxonomy (much 

1967 Wilson & Taylor : Ants of Polynesia < 

synonymy), dist., ecology. 
Chelaner antarcticus : Ettershank, 1966, Austr. J. Zool. 14: 93. 

DISTRIBUTION : KERMADEC IS. : Raoul, records dated 1907-1908 and 1956 (Taylor 
& Wilson 1961). AUSTRAL IS. : Rapa : type locality of rapaense (see above). E. ridge 
of Mt. Perapu, 400-550 mm (ECZ). NE slope of Mt. Tevaitahu, 200-250 m (ECZ). S 
slope of Mt. Tepiahu, 150-200 m (ECZ). 

Fig. 47. Chelaner antarcticus (White), syntype worker of synonymous M. rapa- 
ense Wheeler. 

The rapaense types are about as large as the average worker of New Zealand antarcti- 
cus. Their body coloration falls within the extreme range of antarcticus, being somewhat 
lighter than average and lacking the gastric banding conspicuous in many New Zealand 
samples. The body and appendages are nearly concolorous light reddish brown, the ce- 
phalic dorsum and gastric tergites being a shade darker than the rest of the body. The 
propodeal dorsum is broadly and evenly convex, departing from the flattened condition 
typical of most antarcticus but within the extreme range of the New Zealand population. 
Similarly, the petiolar node is thicker and with a somewhat more convex anterior face 
than the New Zealand average but still not beyond the extreme of variation. In other 
morphological characters rapaense conforms to antarcticus. 

No nest series of New Zealand antarcticus has yet been collected that matches the ra- 
paense types precisely in all characters, yet particular samples can be demonstrated that 
match them in each of the characters described above. If Brown (1958) is followed in 
conceiving antarcticus as an exceptionally variable species, embracing such extreme forms 
as the "species" nitidum Forel and suteri Forel, then rapaense is logically considered con- 
specific. Should new evidence arise to indicate the presence of multiple species in an 
antarcticus complex on New Zealand, the status of rapaense will have to be reconsidered. 

In sum, the present evidence suggests that the Rapa population contains but a small 
segment of the extensive genetic variation of the New Zealand antarcticus population. It 
is interesting to note that antarcticus also occurs on the Chatham and Kermadec Is. Sig- 
nificantly, workers from the latter islands present in the MCZ collection are very similar 
to the rapaense types, differing only in their much smaller size. 

Antarcticus is limited to New Zealand and the Pacific Islands just mentioned. Accord- 
ing to W. E. Moore (in Brown, 1958), the habits of the New Zealand population are 
highly diversified. Colonies nest in rotting logs, under logs and stones, and in crater 
nests in the open soil. They occur in the warm forests of the north, cold bogs at higher 

Pac. Ins. Mon. 1 

altitudes in the south, and open pastures everywhere. The workers are general feeders, 
preying on small insects, scavenging and tending homopterans. On Rapa E. C. Zimmer- 
man collected workers during the day by beating shrubs and low trees. 

Monomorium destructor (Jerdon) 

Fig. 48. 

Atta destructor Jerdon, 1851, Madras J. Lit. Sci. 17: 105. 

DISTRIBUTION : SAMOA: Savai'i : Vaisala, coastal village, strays from house and 
surroundings, banana case shed, and coffee fermenting bin (RWT aces. 466-468). COOK 
IS.: Rarotonga (G. P. Wilder 1925). SOCIETY IS.: Tahiti (Wilder, 1925). *HAWAII : 
Kami: Kekaha. Oahu : Honolulu. Kaimuki. Laysan, 1923 (Fullaway) (Timberlake 1926). 
Intercepted at Honolulu Quarantine Station, from Japan. We have also seen a series from 
Pearl I., Wake Atoll, collected by,N.L.H. Krauss in 1957. 

Fig. 48. Monomorium destructor (Jerdon), worker from United States. 
(After A. D. Cushman in M. R. Smith, 1947). 

This is a pantropical tramp species which probably originated, like other members of 
the subgenus Parholcomyrmex, from Africa. Its distribution in the Pacific is very spotty. 
In Melanesia it is known only from several recent collections from Port Moresby and 
Aroa Plantation, Papua (J. H. Barrett 1957). 

Monomorium floricola (Jerdon) 

Fig. 49. 

Atta floricola Jerdon, 1851, Madras J. Lit. Sci. 17: 107, queen. 

Monomorium speculare Mayr, 1866, Sitz. Akad. Wiss. Wien, 53 : 509 (Type locality : Upolu). 
DISTRIBUTION : SAMOA : Savai'i : Falealupo, dry lowland rain forest, strays on 
foliage, 20.111.1962 (RWT ace. 424). Fanga (NK, 1955). Ga'utavai, coastal coconut grove, 
strays on coconut trunks, 19.111.1962 (RWT ace. 379). Mt. Matafa, 700 m, strays on road- 
side weeds near rain forest, 23.111.1962 (RWT ace. 532). **Safune (1924). Salelologa, 
in domestic gardens, strays on foliage, 23.111.1962 (RWT ace. 547). Upolu : (No further 
data, Mayr 1866). **Apia (1924). Mafa Pass Road (NK, 1955). **Lalomanu (1924). 
Mo'ata, near Apia, numerous colonies nesting in twigs in mangrove trees, strays foraging 
during day, 18.111.1962 (RWT aces. 548, 552, 554, 555, 557-559, 561). **Malololelei (1924). 
Tapatapao, 300m (OHS, ECZ, 1940). Tutuila: west side, Afono Trail, 150m (ECZ 1940). 

Wilson & Taylor : Ants of Polynesia 

Fig. 49. Monomorium floricola (Jerdon), worker from Malaeini, Tutuila (GE). 

**Amouli (1923). Breaker Point (C. Hoyt 1953). Fagatoga, 300 m, beating shrubbery and 
dead branches (ECZ 1940). Malaeini, rain forest, berlesate, moss on rotten log, 13.VI.1962 
(GE ace. T24). Mt. Pioa, 300 m (OHS 1940). **Pago Pago (1923). Utulei, 150-200 m, 
beating vegetation (ECZ 1940). **Manua: Ta'u (1923). Mayr (1866) described the syn- 
onymous M. speculare on the basis of early specimens originating from Upolu. TON- 
GA : Mayr (1876) recorded specimens in Godeffroy Museum of Hamburg. TOKELAU IS.: 
Fakaofo (EHB 1924). COOK IS.: Rarotonga (G. P. Wilder 1925). LINE IS.: Fanning I. 
(S. C. Ball, 1924). *SOCIETY IS.: Bora Bora: W. slope of mountain W of Mt. Pahia, 
200-250 m (1934). Moorea: (about 1924; also 1934). Raiatea: valley above Utura 
(Cheesman & Crawley 1928). Tahiti: numerous lowland localities, 1934, recorded by 
Wheeler (1934b), also recorded by Cheesman & Crawley (1928). *AUSTRAL IS.: Raiva- 
vae: nr. Unurau (1934). Rimatara: Oromana Hills, 80m (1934). Maria I.: NE islet 
(1934). *TUAMOTU IS.: Hao : Boring Bay (1934). South Marutea (1934). *GAMBIER 
lS.:Timoe (1934). Agakauitai (1934). Taravai : NW side (1934). ^MARQUESAS : Ua Hu- 
ka: Tauheeputa, 550m (1928). Eiao : 350-550 m (1928). Hatutu: 450m (1928). *PIT- 
CAIRN I. : (1934). *HENDERSON I. : (1934). *FLINT I. : (1934). HAWAII : Per- 
kins (in Forel 1899) recorded floricola from Oahu, Molokai, Lanai, and Maui and ex- 
pressed the opinion that it occurred on all of the islands. Wheeler (1934a) recorded it 
from localities on Oahu, Maui, Nihoa, and Wake; while Butler (1961) states that he col- 
lected workers and males on Laysan I. both in 1959 and in September 1961. Records 
dated 1923 from Midway and Wake are given by Timberlake (1926). 

M. floricola is one of the most abundant and widespread of all pantropical ant spe- 
cies. Emery (1921) in a review of this and other members of the minutum group, con- 
cluded that floricola originated ultimately from tropical Asia. It is almost wholly arbo- 
real, forming large colonies in trees and bushes in habitats of various degrees of disturb- 
ance. It is a prominent urban species in most tropical countries. Colonies seem unable 
to penetrate undisturbed native forests. 

Monomorium fossulatum Emery 

Fig. 50. 

Monomorium fossulatum Emery, 1894, Ann. Mus. Stor. Nat. Geneva 34: 465-66, worker, 

queen (Type locality: Rangoon, Burma). 
Monomorium fossulatum sechellense Emery, 1894, Ann. Soc. Ent. Fr. 63 : 69-70, text-fig., 

worker, queen (Type locality: Marie-Anne, Seychelles). New Synonymy. 

Pac. Ins. Mon. 

Fig. 50. Monomorium fossulatum Emery, worker from Afiamalu, Upolu (ECZ). 

DISTRIBUTION: SAMOA: Upolu : Afiamalu, 700 m, under dead bark (ECZ 1940). 
*HAWAII : Widespread, recorded from numerous localities on Maui, Kanai, and Oahu. 

Emery's subsp. sechellense is said to differ from " typical " fossulatum by having smaller 
cephalic punctures (these are reported to approach the size and appearance of foveolae 
in fossulatum) and lacking interpunctural sculpturing. In the worker material at our dis- 
posal, all originating from islands in the Pacific and Indian Oceans, this character is 
somewhat variable but does not embrace the extreme fossulatum condition. However, 
Emery states that the queens of the 2 type series are identical, and since fossulatum is 
also so well removed from the remainder of Monomorium, the conspecifity of the 2 forms 
seems probable. 

Fossulatum has been collected in Burma, Formosa, and Papua, as well as in the Sey- 
chelles and Pacific Islands. A related species, subcoecum Emery, has been recorded from 
the West Indies and southern Australia. Taken at face value, the limited evidence seems 
to suggest that fossulatum is native to tropical Asia and includes in its variation a weak- 
ly sculptured form {sechellense) that has been distributed to a limited extent by com- 
merce. The single collection from Afiamalu, Samoa, in 1940 is curious. The species was 
not uncovered by the intensive work by Ettershank & Taylor at the same locality in 1962. 

Monomorium gracillimum (Fr. Smith) 

Myrmica gracillima Fr. Smith, 1861, J. Proc. Linn. Soc. Lond. 6: 34, worker. 


We have been unable to find any reliable differences between African and Indo-Mala- 
yan material separated as destructor and gracillimum respectively by previous workers in 
the Museum of Comparative Zoology. Since destructor (q.v.~) has also been recorded from 
Laysan I. it seems safest to regard Wheeler's determination of a second species, gracillim- 
um, as being in error. 

Monomorium latinode Mayr Fig. 51. 

Monomorium latinode Mayr, 1872, Ann. Mus. Stor. Nat. Genova 2: 152, worker. 

DISTRIBUTION: HAWAII: Oahu : Nuuanu Vail. (Ehrhorn 1923). 

This distinctive species has a native range extending from Ceylon to Formosa and S to 

Wilson & Taylor : Ants of Polynesia 

Fig. 51. Monomorium latinode Mayr, worker from Semarang, Java. 

Java and Sumatra. 

Monomorium minutum Mayr 

Fig. 52. 

Monomorium minutum Mayr, 1855, Verb. Zool.-bot. Ges. Wien 5: 453, worker. 

Monomorium minutum var. liliuokalanii Forel, 1899, Fauna Hawaiiensis (Heterogyna : For- 
micidae), p. 119, worker (Type locality: Honolulu, Hawaii). New Synonymy (Ten- 

Monomorium minutum var. samoanum Santschi, 1928, Insects of Samoa 5 : 46, fig. 26, work- 
er (Type locality: Apia Samoa). New Synonymy (Tentative). 

Fig. 52. Monomorium minutum Mayr, worker from Apia, Upolu (ECZ). 

DISTRIBUTION: SAMOA: Upolu: Apia (ECZ, 1940). **Malololelei, 600 m (1924). 
Tutuila: **Amouli (1923). **Fagasa (1923). **Leone Road (1923). **Pago Pago (1924). 
N. side, Pago Pago, seashore (ECZ, 1940). SOCIETY IS.: Moorea : Baie de Cook (NK 
1955). *Tehau Point (1934). *Tahiti: Tuaruru Vail. Vallee de la Reine. Mataiea. Pa- 
peari. Valley E of Pirae, Pare (1934). MARQUESAS : Ua Huka : Haavei Vail. *HA- 
WAII : Oahu : Honolulu. Manoa and Barbers Point. Kaimuki. Koko Head. Kauai. 
Hawaii: Hilo. Laysan. Midway and Necker (Timberlake 1928). 

Minutum is distributed from Cape Colony to southern Europe, thence sporadically 
through southern and eastern Asia to the Pacific Islands. It has also been recorded from 
Brazil. In the Pacific it is chiefly limited to the most disturbed habitats. 

In the examination of extensive Pacific material, we have been unable to distinguish 

68 Pac. Ins. Mon. 14 

forms that can be considered specifically distinct from the African minutum. Characters 
in the shape and size of the clypeal teeth and length of pubescence used by Forel to sep- 
arate the variety liliuokalanii, and in the shape of the mesosoma and node used by Sants- 
chi to erect samoanum are evidently within the broad range of variation conventionally 
recognized as being embraced by the single species minutum. However, the possibility that 
we are dealing with a group of sibling species cannot be excluded until more extensive 
collections, including queens, are made available from many parts of the range. 

The relatively numerous records from Samoa in the 1920's and in 1940, contrasted with 
the failure of collectors in the late 1950's and in 1962 to capture a single specimen, is 
indicative of considerable fluctuation in population density. It is even possible that the 
species is now extinct in Samoa. 

Monomorium pharaonis (Linne) Fig. 53. 

Formica pharaonis Linne, 1758, Syst. Nat. 10th ed. 1 : 580. 

DISTRIBUTION : SAMOA : Savai'i : Gagaifoumauga, rain forest, berlesates, fern roots 
from rocks and trees and moss on rotting log, 14.VI.1962 (GE ace. SI, S3). **Sala'ilua 
(1924). Salelologa, strays from house and environs (RWT ace. 547). Upolu: Afiamalu, 
700m, disturbed rain forest, berlesate, ground moss, 9.IV.1962 (GE ace. 50). Apia (NK, 
1955). Malololelei, moss in rain forest (TEW, 1956). Le Mafa, 400 m, rain forest, ber- 
lesate, epiphytic fern bases from trees, 13.V.1962 (GE ace. 66). Saleapuga, leaf mold in 
coastal Futu forest (TEW, 1956). Tiapapata, rain forest, berlesates from moss on rotting 
log and from soil, 19.V.1962 (GE aces. 71, 72, 75, 76). Poutasi, leaf mold in rain forest 
(TEW 1956). Above Salani Falls, moss in rain forest (TEW 1956). Tanamalala, leaf 
mold in rain forest (TEW 1956). Falepuna, edge of swamp, leaf litter (TEW 1956). Ma- 
tautu, leaf litter under Ficus (TEW 1956). Vaipoto, rain forest, soil and leaf litter 
(TEW 1956). Tutuila: Pago Pago (C. P. Hoyt 1953). Malaeimi, rain forest, berlesate, 
moss on rotting log, 13.VI.1962 (GE ace. T23); moss on coconut trunk (TEW, 1956). Ta- 
funa (J. L. Gressitt 1952) **Center of island (probably in 1920's). Manua: Ta'u: Lu- 
ma (Wray Harris 1957). Samoa: no further locality (Mayr 1876). TONGA: Tongata- 
bu (W. Cottrell-Darmer, probably sometime before 1935). Tonga : no further locality 
(Mayr 1876). ^SOCIETY IS. : Moorea (1906-07). * MARQUES AS : Ua Huka : Haavei 
Vail., sea level (in 1920's). LINE IS. : Johnston (1947-52, fide Chilson, 1953); before 1934 

Fig. 53. Monomorium pharaonis (L.), worker from Apia, Upolu (NK). 

Wilson & Taylor: Ants of Polynesia 

(*1934b). HAWAII: Wheeler (1934a) records collections from Oahu, Hawaii, French 
Frigates Shoal and Wake. 

This little species is probably the most widely distributed of all ant species. In the 
tropics it occurs in both disturbed and (on Samoa) natural habitats out of doors, as well 
as in human dwellings. In temperate zones it is a prominent house pest around the 
world. Emery, in his 1921 review of Monomorium, suggests that M. pharaonis is native 
to tropical Asia. 

Monomorium talpa Emery 

Fig. 54. 

Monomorium talpa Emery, 1911, Nova Guinea. Zool. 9: 252-53 (Type locality: Etna Bay, 

W. New Guinea). 
Ireneidris myops Donisthorpe, 1943, Ent. Mon. Mag. 79: 81-82 (Type locality: Camp Nok, 


Fig. 54. Monomorium talpa Emery, worker from Utumapu, Upolu (TEW). 

DISTRIBUTION : SAMOA : Savai'i : Gagaifomauga, rain forest, berlesate, leaf mold 
and soil around base of tree, 14.VI.1962 (GE ace. S4). Matega, W. coast, 60 m, banana 
clearing at edge of rain forest, nest in rotting log, workers with dealate queen, 20.111. 
1962 (RWT ace. 397). Mt. Matafa Road, 250 m, rain forest, nest under log, workers and 
2 dealate queens, 19.111.1962 (RWT ace. 369). Mt. Matafa, 650 m, rain forest, nest under 
stone, workers and 2 dealate queens 23. III. 1962 (RWT ace. 544). Mt. Olomanu, rain for- 
est, berlesates of moss on trees, 15.VI.1962 (GE aces. S7, Sll). Upolu: Lauli'i, N. coast, 
1.1956 (TEW). Matautu, berlesate of leaf litter under Ficus (TEW, 1956). Togitogiga, 
400 m, rain forest, berlesates, leaf litter and debris on rotting log, 12. VI. 1962 (GE aces. 
57-59). Utumapu, 130 m (TEW, 1956). Vaipoto, leaf litter and soil in rain forest (TEW, 
1956). Tutuila: Alega, W. Dist., leaf litter and soil in coastal Futu grove (TEW, 1956). 
Tafuna, berlesate of leaf litter in rain forest (TEW, 1956). 

M. talpa is an aberrant member of the genus now known from Waigeo, eastern New 
Guinea, Micronesia, and the Solomons, as well as Samoa. The Samoan material shows 
consistent minor differences from the Melanesian : the mandibles are somewhat less 
slender, the gap between the basal and antebasal mandibular teeth is narrower, and the 
clypeal carinae are slightly more distinct. In these characters the Samoan population is 
more " primitive ", bridging some of the gap between talpa and the rest of Monomorium 

Pac. Ins. Mon. 1 

and making Donisthorpe's Ireneidris less tenable. We have not considered the differences 
between the Melanesian and Samoan forms sufficient to justify recognizing the latter as a 
distinct species, but this is admittedly a borderline case. 

Triglyphothrix striatidens (Emery) Fig. 55. 

Tetramorium obesum subsp. striatidens Emery, 1889, Ann. Mus. Stor. Nat. Geneva 27: 501, 

DISTRIBUTION: SAMOA: Upolu: "Apia, 1000 ft" (Harold Swale, no date). DAN- 
GER IS.: Motu Kotawa (EHB, 1924). LINE IS.: Fanning (A. R. Foster, 1920). 

Fig. 55. Triglyphothrix striatidens (Emery), worker from United States. 
(After A. D. Cushman in M. R. Smith, 1947). 

This species is probably native to tropical Asia and has been spread by commerce to 
Africa, Queensland, and the Pacific coast of Mexico (Taylor & Wilson 1962). The above 
records are the first from Polynesia. The Samoan record is based on a single worn spe- 
cimen taken many years ago ; the species has not been recorded from the islands since, 
despite intensive collecting. T. striatidens was one of the 3 ant species occupying Clip- 
perton Is., in the extreme eastern Pacific, during 1958 ; repetitive collections made by A. 
M. Peterson and C. F. Harbinson indicate it was abundant at that time (Taylor & Wil- 
son 1962). Since the workers are conspicuous and easily collected, the absence of the 
species from collections from most of the Pacific islands probably indicates that it is 
truly very rare or absent through most of this region. 

Tetramorium caespitum (Linne") 

Formica caespita Linne, 1758, Syst. Nat. 10th ed. 1: 581, worker. 

DISTRIBUTION : SOCIETY IS. : Wheeler (1936) mentions a single worker of this 
European species collected by G. P. Wilder at Lake Vaihiria, Tahiti, on 5. V.I 927. Except 
for specimens intercepted sporadically at the Quarantine Station, Honolulu, the species 
has not otherwise been recorded from the Pacific Region. We have not been able to ve- 
rify Wheeler's record. 

Wilson & Taylor : Ants of Polynesia 

Tetramorium guineense (Fabricius) Fig. 56. 

Formica guineense Fabricius, 1793, Ent. Syst. 2: 357, worker. 

Tetramorium guineense var. macro : Santschi, 1928, Insects of Samoa 5 : 50. (l=maera 

Emery, 1914, from New Caledonia). 

DISTRIBUTION: SAMOA : Savai'i: **Fagamalo (1925). Lotogo, 350m, rain forest, 
nest in rotting log, 21.111.1962 (RWT ace. 477). Matega, 65 m, banana clearing at forest 
edge, nest from piece of rotting wood under vegetable debris, and strays from ground 
layer, 20.111.1962 (RWT aces. 400, 409). Palauli (NK, 1955). Salailua, nest under bark 
of coconut tree, 19.111.1962 (RWT ace. 385). Samalaeulu, coastal rain forest, strays under 
small tree-trunk epiphytes, 22.lll.1962 (RWT ace. 509). Upolu: Afiamalu, 700 m, workers 
and winged queens, beating shrubbery, 13.VI.1940 (ECZ ; **1923). Apia (ECZ, 1940; 
**1924). Falefa (NK, 1955). **Malololelei, 650 m (1924). Malololelei Road, 400 m, 
under dead bark (OHS, 1940). Sa'agafou, lowland rain forest, several nests under small 
tree-trunk epiphytes 1-2 m above ground level, 28.111.1962 (RWT ace. 610). Tapatapao, 
under dead bark (OHS, 1940). Tapueleele (NK, 1955). **Tongatata (1923). **Tuaefu 
(1923). **Vaea, 130m (1925). **Vailima, 130m (1925). Tutuila : Afono Trail, west side, 
150 m (ECZ, 1940). Amanave, 130 m (OHS, 1940). **Amouli (1925). Fagaitua (EHB 
1935). Fagatogo, 300 m, beating vegetation (ECZ 1940). Pago Pago (**1925 ; *1931 ; EHB 
1935; NK 1941). Pago Pago, northside 300m, under bark (OHS, ECZ 1940). Pago-Ma- 
tafao Trail, 400m, beating shrubbery (ECZ 1940). Mt. Pioa, 300m (OHS 1940). Manua: 
**Ta'u (1923). Rose (Wray Harris 1938). WALLIS : Emery (1914a) : date of collection 
not given. TONGA: Tongatabu : Nukualofa (NK 1956). Vava'u: Pangai (NK, 1956). 
Falevai (NK, 1956). KERMADEC IS: Raoul, Meyer, records dated 1956 (Taylor & Wilson 
1961) COOK IS.: Aitutaki: Akaiami (NK, 1955). TOKELAU IS.: Fakaofo (EHB, 
1924). PHOENIX IS.: Canton (EHB 1924; Wilson 1954; NK 1958); 1940 fide van 
Zwaluwenburg (1943). *DANGER IS.: Puka Puka (1933). LINE IS.: Palmyra (EHB 
1938). E. J. Ford (1959) records this species on sprouted coconuts brought from Palmyra 
to Honolulu in 1957. Christmas (L. A. Whitney 1924). Johnston (1947-52), fide Chilson 
(1953). SOCIETY IS.: *Tahiti (1926-34). *Bora Bora (1934). Raiatea (*1934). Opoa 
and Uturoa (NK, 1955). Cheesman & Crawley (1928) : Tahiti (1925), Raiatea (1925). 
AUSTRAL IS.: *Maria: NE islet (1934). Rimatara (A. M. Stokes 1921; ECZ 1934). 
*Raivavae (1934). Rapa (A. M. Stokes, 1921; ECZ, 1934). TUAMOTU IS.: *Tepoto 

Fig. 56. Tetramorium guineense (Fbr.), worker from Upolu (ECZ), 

Pac. Ins. Mon. 

(1934). *Timo: north islet (1934). Makatea (G. P. Wilder 1932). GAMBIER IS.: 
*Taravai (1936). *Mangareva (1934). MARQUESAS : *Hiva Oa (1929). *Tahuata (1929). 
*Fa/ #iva (1929). *Mohotane (1929). */a #fca (1929). *Eia0 (1929). **Hatutu. Chees- 
man & Crawley: Nuku Hiva (1925). *PITCAIRN (1934). *HENDERSON (1934). 
*OENO (1934). EASTER (*1922) : The species was collected at the following places by 
I. E. Efford and J. A. Mathias in I-II.1965: in Zea mais, Hanga Roa ; on grass, hillside 
behind beach at Ovahe ; "litter under Eucalyptus village" ; in crater, Rano Raraku. Wing- 
ed queens and $$ were collected on 26 & 29. I. *HAWAII : Oahu, Kami, Maul, Molokai, 
Hawaii. Timberlake (1926) lists records made in 1923 on the following islands: Nihoa, 
Necker, Midway, Laysan, Pearl and Hermes Reef, Kure (Ocean). We have seen material 
collected on Laysan and Kure by G. D. Butler in IX.1961. Alate queens are present in 
the Laysan sample. 

Guineense is one of the most abundant and widespread of the pantropical tramp species. 
It is evidently native to Africa, where it is associated with related endemic species and 
shows considerable infraspecific variation of its own. It is abundant throughout tropical 
Asia in disturbed habitats and ranges into Queensland and through the pacific. In the 
New World it occurs in many disturbed habitats as a dominant ant, from South America 
through the West Indies to the Florida Keys, and it is also occasionally encountered in 
greenhouses in various parts of the United States. The colonies are mostly arboricolous, 
while the workers forage both arboreally and on the ground. 

Tetramorium pacificum Mayr Fig. 57. 

Tetramorium pacificum Mayr, 1870, Verh. Zool.-bot. Ges. Wien 20: 38, worker, (Type lo- 
cality: Tongatabu, Tonga). 

DISTRIBUTION : SAMOA : Savai'i : Le Maf a, roadsides, strays under moss on tree 
trunk (RWT ace. 635). Lotogo, 370 m, rain forest, 3 nests in hollow twig on Ficus tree, 
1-35 m above ground (RWT ace. 457, 458, 460) ; strays on foliage (RWT ace. 451). Mt. 
Matafa, 700 m, rain forest ; strays in leaf mold at base of tree fern (RWT ace. 520) ; 
nest in hollow twig on tree (RWT ace. 524) ; nest under small tree-trunk epiphyte (RWT 
ace. 531); strays on roadside weeds (RWT ace. 532). Mt. Matafa Road, 200m, rain 
forest, strays on foliage (RWT ace. 361). Puapua (NK, 1955). Palauli (NK, 1955). 
**Salailua (1924). Upolu: Afiamalu, 700m, beating shrubbery (ECZ, 1940); disturbed 
forest, colony in rotting tree stump (RWT ace. 314) ; dealate queen, stray on tree trunk 
(RWT ace. 343) ; sweeping roadside weeds (RWT ace. 252). Mafa Pass Road (NK, 1955) 
Malololelei (OHS 1940; NK 1955). Sa'agafou, lowland rain forest, colony in hollow twig 
on tree (RWT ace. 610). Sinaele, 400m, beating vegetation (ECZ, 1940). **Taefu (1924). 
Tapatapao, 250 m, beating vegetation (ECZ 1940). Tapueleele (NK 1955). Tanumalala, 
350m, disturbed weed-overgrown rain forest, strays in foliage (RWT ace. 606). Tutuila: 
Amanava, 130 m (ECZ, OHS, 1940). Afono Trail, W. side, 130 m (ECZ 1940). Amouli, 
130m (**1923; D. T. Fullaway 1930; ECZ 1940). Fagasa (D. T. Fullaway 1930). Fa- 
gasa Trail, 200 m (OHS, 1940). Fagatogo, 300m beating shrubs (ECZ, 1940). Fangatanga, 
Reservoir, beating shrubs (ECZ 1938). Leone-Aula (D. T. Fullaway 1930). **Leone Road 
(1923). Moloata, 300m, beating dead branches (ECZ, 1940). **Pago Pago (1924). Pago 
Pago, N. side, 400m, beating dead branches (ECZ 1940). Mt. Pioa, 300m (OHS 1940). 
**Manua : Ta'u (1923). All Samoan collections by RWT were made in III-VI.1962. TONGA : 

Wilson & Taylor : Ants of Polynesia 

Tongatabu: Nukualofa (NK 1956). Record by Mayr (1870). Vava'u: Neiafu (NK 1956). 
Eua : Ohonua (NK, 1956). ROTUMA : Solkope (H. St. John 1938). COOK IS. : Rarotonga 
(G. P. Wilder 1927, 1929). SOCIETY IS. : Tahiti: Papenoo Vail. NK, 1955) ; (*1927, 1934 ; 
Cheesman & Crawley 1925). *Moorea (1924). *Mehetia (1934). *Raiatea (1934). *AU- 
STRAL IS.: Raivavae (1934). Rautaro Islet, Tubuai (1934). Rurutu (1934). Rapa (1934). 
*TUAMOTU IS. : Makatea (1932). *GAMBIER IS. : Mangareva (1934). *MARQUESAS : 
Fatu Hiva. Nuku Hiva. 

T. pacificum ranges over most or all of f O^'uJX'f K\\ // 

SE Asia, most of Melanesia, including J' ^-iUiXNV , ., .//2 

New Caledonia, Micronesia, and Poly- ^<^ J \ if J Y^ ^T^- 

nesia east to the Marquesas. The species J"^a^\ N 

is perhaps native to most or all of this / 

range. On the other hand, pacificum can 57 

be carried by human commerce. Individ- ., . 

uals have been intercepted in quarantine ^x^ \^ l \Ww\\K/l ,\i, /. . 

at Honolulu, although the species has 

not to our knowledge become establish- \ I f 

ed in the Hawaiian Is. At least 6 re- 

lated species occur in Melanesia, includ- /* 

ing bicolor Viehmeyer, melanogyna Mann, 

validiusculum Emery, and wilsoni Mann. 

On New Guinea pacificum is mostly re- ^ 

placed in natural habitats by the sibling 

species validiusculum. We have been un- ^V V. \ ^ 

able to find any significant geographical ^ 

variation in the species. / 


Tetramorium simillimum (Fr. Smith) Fig. 57. Tetramorium pacificum Mayr, worker 

pig. 59. from Afiamalu, Upolu (ECZ). Fig. 58. Tetra- 

,, . . .,,. TJ, o~.; t u loci T ;ct morium tonganum Mayr, worker from N. Side, 
Myrmica simillima Fr. smith, Io51, List s J ' . _, 

' Pago Pago, Tutuila (ECZ). Fig. 59. Tetrama- 

Brit. Anim. Bnt. Mus. 6: 118, work- rium simi u imum (p r . Smith), worker from Jarvis 
er. I., Line Is. (W. W. Graf). 

DISTRIBUTION : SAMOA : Savai'i : Falealupo, dry rain forest, nest under bark of dry 
log, 20.111.1962 (RWT ace. 426). Ga'utavai, coastal coconut grove, strays on palm trunks, 
19.111.1962 (RWT ace. 379). Lotogo, 350 m, strays in banana plantation, 21.111.1962 (RWT 
ace. 469). Salelologa, in domestic garden, 23.111.1962 (RWT ace. 547). Upolu : Apia (H. 
Swale, no date, probably 1920's or 1930's). Tongatabu (W. Cottrell-Dormer, no date). 
COOK IS.: Rarotonga (1937). TOKELAU IS.: Fakaofo (EHB, 1924). PHOENIX IS.: 
Phoenix (EHB, 1924). Howland (195) Canton, in soil beneath heaps of old coconuts, 
1940-41 (fide van Zwaluwenburg, 1943). LINE IS. : Jarvis, bunch grass (W. W. Graf, 1935). 
SOCIETY IS.: Cheesman & Crawley (1928): Tahiti (1925), Bora Bora (1925). *AUSTRAL 
IS. : Raivavae (1934). EASTER IS. : (*1922) ; at camp, Hanga Roa, 4.II.1965 (I. E. Efford 
& J. A. Mathias). *HAWAII : Oahu. 

Like its larger and better known congener T. guineense, T. simillimum is native to Africa 
and has been widely distributed through the tropics by commerce. 

Pac. Ins. Mon. 

Tetramorium tonganum Mayr Fig. 58. 

Tetramorium tonganum Mayr, 1870, Verb. Zool.-bot. Ges. Wien 20: 38, worker (Type lo- 
cality: Tongatabu, Tonga). 

DISTRIBUTION: SAMOA: Savai'i: Asau, 170m, lowland rain forest, nest under small 
trunk epiphyte (RWT and GE ace. 491). Ga'utavai, coastal coconut grove, nest under 
bark of coconut palm (RW and GE ace. 383). Gagaifoumauga, rain forest, moss on rot- 
6). Lotogo, 370m, strays on Ficus tree, 35m, above ground (RWT 
and GE ace. 459). Matega, 65 m, banana clearing, forest edge, strays under bark and on 
surface of rotting log (RWT aces. 396, 401) ; disturbed rain forest, strays on ground and 
in litter (RWT and GE ace. 402). Patamea (NK, 1955). Puapua (NK, 1955). **Sala'- 
ilua (1924). Samalaeulu, coastal rain forest, one colony and several series of strays under 
small trunk epiphytes (RWT and GE aces. 509, 510, 515). Upolu : Afiamalu, 700m, dis- 
turbed forest, berlesates, ground moss (GE ace. 53). Le Mafa, 400m, rain forest, berle- 
sates, leaf litter, soil, moss and epiphytes (GE aces. 61, 63, 66). Malololelei Road, 500-600 
m, beating shrubbery (ECZ, 1940; NK, 1955). Mafa Pass Road (NK, 1955). Tiapapata, 
rain forest, berlesates, moss on trees, vines, and epiphytes (GE aces. 71, 72, 74). **Vailima 
(1924). Tutuila: N of Aloau, 430m, stunted forest, berlesates, moss on logs, with alate 
queens, 13.VII.1962 (GE aces. T16, T18). Amouli, 160 m, beating (ECZ, 1940). **Fagasa 
(1923). Fagatogo, 200 m, beating shrubbery (ECZ 1940). Reservoir, Fangatanga (ECZ, 
1938). Le Mafa, 250 m, rain forest, berlesate, moss on rotting log (GE ace. T9). **Leone 
Road (1923). Malaeini, stunted forest, berlesate, moss on rotting log (GE ace. T21). 
Matafao Trail, 100 m, beating dead branches (ECZ, 1940). Maupasaga, Fagasa Trail (D. 
T. Fullaway 1930). Pago Pago, N. side, 400 m, beating dead branches (ECZ 1940). Pago 
Pago (EHB, 1935; NK 1941; **1924). Between Oloaua and Olomanu, rain forest, ber- 
lesates, and moss on trees and epiphytes (GE aces. Til, T12). All Samoan records 
by Ettershank & Taylor from March through July 1962. TONGA: Mayr (1870).* LINE 
IS. : Christmas (1934). ^SOCIETY IS. : Huahine (1934). Moorea (1934). Raiatea (1934). 
Mehetia (1934). *AUSTRAL IS.: Rimatara (1934). Rurutu (1934). MARQUESAS: 
*Hiva Oa (1929). *Fatu Hiva. Cheesman & Crawley (1928) : Nuku Him (1925). HA- 
WAII : *Oahu (1927). Kure Island, September 1961 (G. D. Butler). 

Tonganum ranges more or less continuously from New Britain to Micronesia and eastern 
Polynesia. It is apparently absent from New Guinea, New Caledonia, and Australia. The 
compact, restricted nature of this distribution suggests that it must be natural at least in 
part. This interpretation is supported by the presence on New Guinea of a related endem- 
ic species, wagneri Viehmeyer. Whether tonganum is native to all or part of its current 
range in Polynesia is not ascertainable on the basis of present evidence. 

Rogeria exsulans Wilson and Taylor, new species Fig. 60. 

DIAGNOSIS AND RELATIONSHIPS : This species is a true member of Rogeria s. str., a genus 
of rather rare species hitherto known only from the New World tropics (W. W. Kempf 
1962). We are grateful to Dr Kempf for comparing paratypes with his extensive collection 
of Rogeria and aiding in making the following diagnosis. Exsulans is most similar to 
germaini Emery of Brazil and scabra Weber of Cuba. 

Wilson & Taylor : Ants of Polynesia 


Fig. 60. Rogeria exsulans Wilson and Taylor, paratype worker (Afiamalu, 
Samoa, collected by E. C. Zimmerman). 

It differs from scabra in the following characters: (1) in exsulans the space between 
the posterior margin of the propodeal spiracle and the posterior border of the declivitous 
propodeal face is approximately equal to the maximum width of the spiracle, whereas in 
scabra this space is less than 1/2 the maximum spiracular width ; (2) the subpetiolar flange 
is wider in exsulans, especially in the anterior 1/2 ; (3) the scapes of exsulans are propor- 
tionately longer ; in repose and directed toward the occipital angles they fail to reach 
the border of the angles by approximately their own maximum width ; whereas in scabra 
the scapes fail to reach the borders by nearly 2x their maximum width; (4) in exsulans 
the mesosomal sculpturing is composed of a rugoreticulum with no distinct longitudinal 
orientation ; in scabra the major mesosomal sculpturing is composed predominantly of lon- 
gitudinally oriented rugae. In other details of body form, sculpturing, and color the 2 
species are very similar. The differences cited have been noted to hold consistently for 
extensive series of scabra examined from several localities in central and western Cuba. 
(The 3 Cuban forms brunnea Santschi 1930, scabra Weber 1934, and caraiba Santschi 1936 
are all very close and very possibly conspecific with one another. Cuban material in the 
Museum of Comparative Zoology, Harvard University, gathered about the syntypes of 
scabra is highly and continuously variable in the characters which are supposed to be 
diagnostic for brunnea and caraiba). 

Exsulans differs from germaini by the strictly decumbent and relatively long pilosity of 
the gaster and the completely rugoreticulate sculpture of the thoracic dorsum. 

The above species are in turn related to belli Mann of Central America, cubensis Sant- 
schi of Cuba, and 2 or 3 additional species, probably undescribed, which we have obtain- 
ed from southern Mexico. From all of these exsulans can be easily separated by the com- 
bination of relatively large eye size (EL 0.09 mm) ; relatively small, anteriorly placed pro- 
podeal spiracle ; well developed subpetiolar flange ; moderately large total size ; and dense 
rugoreticulate mesosomal sculpturing. Finally, exsulans is completely distinct from manni, 
stigmatica and sublevinodis, the only other Old World members of Rogeria s. lat. The char- 
acters shown in fig. 60 can be used to separate it from these other species. 

Holotype worker: HW 0.57 mm, HL 0.66mm, SL 0.46mm, length of 1st segment of an- 
tennal club 0.07 mm ; length of 2nd club segment 0.09 mm, length of 3rd (terminal) club 
segment 0.27 mm, length of funiculus exclusive of club 0.24 mm, EL 0.09 mm (eye con- 
tains about 15 ommatidia). 

Pac. Ins. Mon. 

MATERIAL EXAMINED : SAMOA : Upolu : Afiamalu, 700 m, under dead bark, 9.XI.1940 
(ECZ) ; rain forest, from fern and moss on tree trunk, 19.1.1956 (TEW) ; disturbed rain 
forest, berlesates of moss on logs (RWT ace. 2318; GE ace. 54) and berlesates of moss 
on tree trunks 8-9 m from ground (RWT ace. 2308 ; GE ace. 45) ; strays from rotting logs 
and small fragments of rotting wood on forest floor (RWT aces. 223, 282, 339, 601). All 
RWT and GE collections were made in III and IV, 1962. The holotype was drawn from 
the series collected in 1956 by T. E. Woodward. 

As noted in the comparative description given above, exsulans is a distinct species close- 
ly allied to a complex of species endemic to Mexico, Central America, and the West 
Indies. None of these forms is known to have been dispersed by human commerce, and 
there is no basis to assume that exsulans reached the western Pacific in this manner. Fur- 
thermore, exsulans itself is not known from the New World. Until contrary evidence is 
obtained exsulans should be tentatively considered an endemic Samoan species. If this is 
truly the case, it represents one of the most anomalous discontinuous distributions found 
in the ants. A roughly parallel case is seen in the genus Rhopalothrix, which is represent- 
ed by several species in the New World tropics and one undoubted endemic on New 
Guinea (Brown & Kempf 1960). 

Rogeria sublevinodis Emery, new status Fig. 61. 

Rogeria stigmatica subsp. sublevinodis Emery, 1914, Nova Caledonia (Zool.) 1: 415, worker 
(Type locality: Raoua, Mare, Loyalty Is.). 

DISTRIBUTION: SAMOA: Savai'i: Alagaogao, 100m, banana plantation, strays from 
rotting wood on ground (RWT ace. 417). Falealupo, rain forest, from fragment of rot- 
ting wood on forest floor (RWT ace. 439). Gagaifoumauga, rain forest, berlesates of leaf 
mold, soil, and rotting log (GE aces. S4, S5). Lotogo, 350 m, rain forest, nest in frag- 
ment of rotting wood on forest floor (RWT ace. 471). Matega, 65m, edge of rain forest and 
banana plantation, nest under bark of rotting log (RWT ace. 406). Mt. Matafa Rd., 200 
m, rain forest, nest with alates, under moss and bark on rotting log : also strays under bark 
of rotting log, 19.111.1962 (RWT aces. 367, 368). Upolu : Falepuna, leaf litter at swamp 
edge (TEW, 1956). Le Mafa, 400 m, rain forest, berlesate of epiphytic fern bases on trees 
(GE ace. 66). Lauli'i (TEW, 1956). Matautu, from moss under Ficus (TEW, 1956). 
Poutasi, rain forest, leaf mold (TEW, 1956). Sale Puga, coastal Futu forest, leaf mold 
(TEW, 1956). Sliding Rock, 200 m, rain forest, colony in rotting wood fragment on for- 
est floor (RWT ace. 285). Tanumalala, 350 m, disturbed and weed-overgrown rain for- 
est, nest under leaf litter at side of rotting 

/A/ stump (RWT ace. 607). Utumapu, 400m 

*^-- (TEW, 1956); rain forest, nest with alates un- 

y*\ der stones, 17.111.1962 (RWT aces. 567, 569). 

^ Vaipoto, rain forest, leaf litter and soil (TEW, 
fc^ 1956). Tutuila: Alega, coastal Futu grove, leaf 
^ litter and soil (TEW, 1956). N of Aloau, 450 
/ m, stunted forest, berlesate of moss from rot- 

Fig. 61. Rogeria sublevinodis Emery, ting lo S ( GE acc - T2 ) ' Pa S Pa S (collector 
worker from Saleapuga, Upolu (TEW). unknown, 1910). Tafuna, rain forest, leaf 

mold and soil (TEW 1956). (All Samoan col- 

1967 Wilson & Taylor : Ants of Polynesia 

lections by Ettershank & Taylor were made in III-VII.1962). SOCIETY: Tahiti: Riv- 
nac, Punaauia Dist., "berlese open field" (J. M. Dixon 1949; deposited in Chicago Muse- 
um of Natural History). Raiatea: Cheesman & Crawley (1928) : "in the interior of the 
northern end of the island, at the head of a gully at 2000 ft. in dense scrub. Series of 
workers and portion of the nest which was in a rotten log..." 

Sublevinodis occurs throughout the Fiji Is., even as far as Munia, Lau (collected by W. 
M. Mann in 1914). It is also known from the Loyalty Is. (type locality). From these 
parts of outer Melanesia its range extends apparently spottily to the Society Is. and Mic- 
ronesia. It is replaced in New Guinea and the Solomons by stigmatica Emery, which is 
larger and with denser dorsal pedicellar sculpturing. We have been unable to find any 
significant differences among workers from Samoa, Tahiti and Fiji. 

Adelomyrmex (Arctomyrmex) samoanus Wilson and Taylor, new species 

Fig. 62. 

DIAGNOSIS : Similar to the only other described species of the subgenus, Adelomyrmex 
(Arctomyrmex) hirsutus Mann of Fiji and New Caledonia, differing (in the worker caste) 
in having a much more deeply impressed metanotal groove and much heavier sculptura- 
tion, especially on head. 

Fig. 62. Adelomyrmex (Arctomyrmex) samoanus Wilson and Taylor, para- 
type worker from the type locality, Urumapu, Upolu. 

Holotype worker: HW 0.45 mm, HL 0.50 mm, CI 90, SL 0.31 mm, SI 69. Head, anten- 
na, and body form of the very distinctive Arctomyrmex form shown in fig. 62. Mandi- 
bles feebly and longitudinally striate, their surfaces shining. Clypeus smooth and shining; 
almost entire remainder of head rugoreticulate, diameters of round forea-like interspaces 
about 0.01 mm, surface of interspaces shagreened and opaque, surfaces of reticulum itself 
feebly shining. Meso- and metathorax longitudinally rugose and feebly shining, prothorax 
and most of propodeum sparsely punctate and strongly shining. Petiolar node coarsely 
sculptured in a manner similar to most of cephalic surface. Postpetiole and gaster mostly 
sparsely punctate and strongly shining. [A. hirsutus, described from Fiji by Mann (1921) 
and more recently collected at several localities on New Caledonia by E. O. Wilson, has 
almost the entire surface of the body sparsely punctate and strongly shining.] Most of 
dorsal surface of head medium brown. Remainder of body light brown. Appendages 
clear brownish yellow. 

78 Pac. Ins. Mon. 14 

Paratypes : Workers from Upolu and Tutuila do not deviate significantly from the holo- 
type specimen. The HW of the single Tutuila specimens measured is 0.45 mm ; that of 
the extensive Upolu series ranges 0.42-0.49 mm. The single Savai'i worker, on the other 
hand, is quite different. It is larger (HW 0.60 mm) ; longitudinal rugae extend from the 
mesothorax halfway across the pronotum, and the rugoreticulum of the petiole is more 
shallow than in other Samoan material. 

MATERIAL EXAMINED : SAMOA : Savai'i : Gagaif oumauga, rain forest, berlesate, soil around 
tree roots, 1 worker, 14.V.1962 (GE ace. S4). Upolu: Le Mafa, rain forest, berlesate, 
leaf litter, 1 worker (GE ace. 63). Poutasi, rain forest, leaf mold, 9 workers (TEW 1956). 
Above Salani Falls, rain forest, leaf mold, 1 worker (TEW 1956). Utumapu, 130m, 20 
workers including holotype, 17.1.1956 (TEW). Tutuila: Tafuna, rain forest, berlesate, leaf 
mold and soil, 1 worker (TEW 1956). 

Iridomynnex anceps (Roger) Fig. 63. 

Formica anceps Roger, Berl. Ent. Z. 7: 164-165, worker (Type locality: Malacca, Malaya). 
Iridomyrmex gracilis subsp. papuanus Emery, 1897, Ann. Mus. Civ. Stor. Nat. Genova 18: 

572, worker (Type locality: Kapa Kapa, New Guinea). New Synonymy. 
Iridomyrmex anceps r. papuanus: Forel, 1901, Mitt. Zool. Mus. Berlin 2: 19 worker queen 

Fig. 63. Iridomyrmex anceps (Roger), worker from Akaiami Aitutaki 
Cook Is. (NK). 

DISTRIBUTION: COOK IS.: Aitutaki: Akaiami (NK 1955). 

Anceps is the most widespread of the Indo-Australian Iridomyrmex, ranging continuous- 
ly from India to eastern Australia, through Micronesia and almost all of Melanesia. The 
above record is the first from Polynesia. Evidently the species is in the process of spread- 
ing with the help of man. We have recently seen the first series collected on Fiji ; these 
originate from Nandi, Viti Levu (an international airport) and were taken by E. J. Ford, 
Jr., in May 1956. The great distance of the Cook Is. from the nearest source populations 
also indicates spread by commerce. 

1967 Wilson & Taylor : Ants of Polynesia 

Iridomynnex humilis (Mayr) Fig. 64. 

Hypoclinea humilis Mayr, 1868, Ann. Soc. Nat. Modena 3 : 164, worker (Type locality : 
Buenos Aires, Argentina). 

DISTRIBUTION: HAWAII: Oahu. Hawaii. Maui. 

Before it became established this famous South American pest species was frequently 
intercepted at the Honolulu quarantine station on goods coming from California during 
the 1930's (*1934a; Zimmerman 1940). It finally gained a foothold, probably in the Spring 
of 1940, in Fort Shafter, within the center of Honolulu. Zimmerman (1940) reported 
that when he examined the population about X.I 940, "numerous strongly developed colo- 
nies" were present but all were confined to Fort Shafter. The species appeared to be 
in the act of eliminating other ant species it contacted, a destructive ecological trait it 
has exhibited elsewhere. Pemberton (1944) reported that in 1944 the range of the spe- 
cies occupied a growing area two-fifth by four-fifth kilometer within the military reser- 
vation, but that it was apparently still confined to this one locality. 

Fig. 64. Iridomyrmex humilis Mayr, worker from United States. (After S. 
H. DeBord in M. R. Smith, 1947). 

By September 1949, the species had expanded beyond Fort Shafter. On this date W. 
Thomsen and G. D. Peterson, Jr. (Proc. Hawn. Ent. Soc. 14 : 14, 1950) found colonies at 
Pearl Harbor, at that time "the farthest known extension" from Fort Shafter. By Octo- 
ber 1949, Pemberton and Thomsen (Ibid. 14: 16, 1950) found humilis at Nuuanu Moana- 
lua Gardens, Moanalua golf course, and north Halawa Valley. In November of the same 
year, Thomsen & Pemberton (Ibid. 14 : 20) traced colonies throughout the Ewa Planta- 
tion, "from Pump 10 to the western end." Interestingly, these investigators suggested that 
the original point of infestation was an army camp, since abandoned, in kiawe growth 
nearby. By 1949 the infestation was obviously spreading by jumps, aided by commerce. On 
29 September, 1950, Thomsen (Ibid. 14 : 219, 1951) discovered humilis on Route 212 four- 
fifths kilometer S of Mauna Kapu, in the SE ridge of the Waianae Range, Oahu, at 750 
m elevation. From this center the infestation continued about two-fifths kilometer in ei- 
ther direction along the road. 

In November 1950, populations of humilis were reported from Kamuela, Hawaii, and 
Makawao, Maui, by J. E. Eckert (Ibid. 14 : 222). Their origins were unknown, although 
Eckert suggested that the Maui population, which was in a flourishing condition, was es- 

Pac. Ins. Mon. 

tablished by troops quartered in the area during World War II. 

Subsequent reports by Thomsen, Chilson, K. Ito, L. F. Steiner, and R. H. van Zwalu- 
wenburg in the "Notes and Exhibitions " sections of the Proc. Hawn. Ent. Soc. follow in 
detail the discoveries of additional infestations on Oahu subsequent to 1950. Among the 
more noteworthy records is that of Chilson (1956) who found humilis on 15 March 1955 
at Kaneohe, the first record for windward Oahu. It is clear that, as in other countries, 
the species disperses on its own only very slowly, apparently solely by emigration of co- 
lonies over the ground. Nuptial flights involving queens are rare or non-existent. The 
longer jumps in distribution are made by colonies carried through human commerce. As 
Thomsen has pointed out, many of the Hawaiian local populations are associated with 
army camps and bivouacs, and it is reasonable to assume that the colonies were carried 
inadvertently with supplies and equipment. The species probably became firmly establish- 
ed and widely distributed in this way during World War II and shortly afterward. 

The authors cited confirm what has been observed elsewhere, that /. humilis excludes 
other larger ant species, including the formidable Pheidole megacephala. One species found 
to be compatible with it on Hawaii is the diminutive Cardiocondyla nuda. 

Zatapinoma wheeleri Mann Fig. 65. 

(Doubtful record). 

Zatapinoma wheeleri Mann, 1935, Psyche 42: 36-37, fig. 2, soldier (Type locality: "taken 
in quarantine at Honolulu on plants from Samoa"). 

While wheeleri appears to be a valid species, its presumptive origin in Samoa (taken in 
quarantine at Honolulu) must be doubted until additional material is obtained. When it 

is recalled that Samoa has been so in- 

/^y^~\ ,--T \_^ (^ { I 7~\ tensively collected for ants, especial- 

i /cC V-^r^L 1 1 L^ ly arboricolous species, the complete 

lack of additional verified records ren- 

,, /" \ ders the existence of Zatapinoma in 

/ \fa tne islands improbable. Moreover, no 

species of the genus has been record- 
ed from Melanesia. On the other side, 
i* (J significance may be lent to series of 
^-3L/ winged queens of 2 undescribed spe- 

Fig. 65. Zatapinoma wheeleri Mann. Syntype cies that we have recently seen from 
soldier (after W. M. Mann). the Palau Is. The possibility exists 

that these ants occupy unusual habitats 
that make them inaccessible to usual collecting methods. 

Tapinoma melanocephalum (Fabricius) 

Fig. 66. 

Formica melanocephalum Fabricius, 1793, Ent. Syst. 2: 353, worker. 

Tapinoma (Micromyrma~) melanocephalum var. australis Santschi, 1928, Insects of Samoa 5: 

53, worker (Type locality: "Mog. Narbor", Espiritu Santo, New Hebrides). New 


DISTRIBUTION : SAMOA : Savai'i : Alagaogao, 100 m, near Neiafu, strays on roadside 

Wilson & Taylor : Ants of Polynesia 

Fig. 66. Tapinoma melanocephalum (Fabr.), worker from Opoa, Raiatea (NK, 1955). 

weeds (RWT and GE ace. 416). Falealupo, dry rain forest, strays on ground (RWT and 
GE ace. 424). Lotogo, 370 m, rain forest, strays on Ficus, 35 m above ground level (RWT 
ace. 459) ; banana plantation, strays on foliage (RWT ace. 469). Matega, 65 m, banana 
clearing at edge of rain forest, strays from moss and leaf mold on boulders (RWT ace. 
390) ; rain forest, strays from ground layer (RWT ace. 409). Papa, banana plantation, 
strays on banana plants (RWT ace. 375) ; nest under bark of dry dadap log (RWT ace! 
378). Puapua (NK, 1955). Samalaeulu (NK, 1955). Sili (NK, 1955). **Safune (1924)! 
Upolu: Apia (ECZ, 1940; **1923). Lefanga (NK, 1955). Le Mafa, strays around roadside 
cutting, 29.111.1962 (RWT ace. 635). Mo'ata, near Apia, several nests in hollow twigs in 
mangrove forest (RWT ace. 560). Tanumulala, 350 m, disturbed weed overgrown rain 
forest, strays ex foliage (RWT ace. 610). Tapatapao, 250 m, beating dead branches 
(ECZ, 1940). Tapueleele (NK, 1955). Tutuila: N of Aloau, 450 m, stunted forest, 
berlesate, moss on logs (GE ace. T17). **Amouli (1923). **Leone Road (1924). N. side, 
Pago Pago, 200m, from tree fern (OHS, 1940). Pago Pago (Wheeler 1931 ; **1923). Utulei! 
150-200 m, beating (ECZ, 1940). Recorded from Samoa by Mayr (1876). TONGA: 
Recorded by Mayr (1876). No further data (H. Cottrell-Dormer). TOKELAU IS. : Fa- 
kaofo (EHB, 1924). COOK IS. : Rarotonga. SOCIETY ISLANDS : Moorea: *( 1906^07, 
1925). Baie de Cook (NK, 1955). *Tehau Pt. (1934). Tahiti : *Fautaua, 250 m. "Abun^ 
dant" around Papeete in 1925 (Cheesman & Crawley, 1928). Papenoo Valley (NK, 1955). 
*Valley of Pirae, Pare (1934). *Tuauru. *Urufara Valley, 100 m (1934). Raiatea: Opoa 
(NK, 1955). Uturoa (Cheesman & Crawley, 1928). PHOENIX IS.: Canton, 1940-41, fide 
van Zwaluwenburg (1943). *AUSTRAL IS. : Maria, NE islet (1934). *TUAMOTU IS. : 
Timoe, northern islet (1934). *GAMBIER IS.: Akamaru, N. side (1934). *MARQUESAS : 
Eiao, 500 m. Hatutu, 300 m. Ua Huka : Teanatuhiva, 100 m ; Hanapoua Valley. *HEN- 
DERSON I.: N. side (1934). *FLINT I. (1934). HAWAII: "Honolulu and Waianae, 
Oahu; also on Lanai and Molokai and no doubt the other islands" (Perkins in Forel, 
1899). Emery (1899) records melanocephalum as one of the 4 ant species collected on 
Laysan by Schauinsland in 1896-97. Wheeler (1934a) reported it to be widespread on 
Oahu, Lanai, Molokai, Hawaii, Nihoa, Midway, Ocean I., and Laysan. Clagg (1957) re- 
ports that the species was formerly a household pest in Honolulu but disappeared by 1949. 
Some additional collections were made in the Pearl Harbor area in 1956. In September 
1961, G. D. Butler collected the species on Kure I. 

; Pac. Ins. Mon. 1 

The origin of this ubiquitous tramp species is unknown. Related species are native to 
various parts of Africa, southeastern Asia, and the New World subtropics and tropics. 

Tapinoma minutum Mayr Fig. 67. 

Tapinoma minutum Mayr, 1862, Verh. Zool.-bot. 

s ~~~V? Ges - Wien 12 : 703 ' worker (TyP e locality: 

I // New South Wales). 

Prenolepis minutula atomus var. fullawayi Wheel- 

lr\ llnl \^ [ I \\ er, 1912, J. N.Y. Ent. Soc. 20: 46, worker 

"\ \ \\ (Type locality: Guam). New Synonymy. 

_, DISTRIBUTION: SAMOA: Savai'i: Falea- 

_>. "' lupo, dry forest, strays from foliage (RWT ace. 

424). Upolu : Afiamalu, 700 m, beating (ECZ, 
Fig. 67. Tapinoma minutum Mayr, work- 1940) ; disturbed rain forest, strays on ground 
er from Afiamalu, Upolu (ECZ). and under feark of log (RWT accs 211< 341) 

Mo'ata, coastal mangroves, nest with gff in dead twig, 18.111.1962 (RWT ace. 558). 

We have seen additional material of this inconspicuous little species from Queensland, 
New Guinea, Solomons, Fiji, and Micronesia. The new synonymy is based on examina- 
tion of Wheeler's types of fullawayi. 

Technomyrntex albipes (Fr. Smith) Fig. 68. 

Formica (Tapinoma} albipes Fr. Smith, 1861, Proc. Linn. Soc. London, Zool. 6: 38, worker. 

(Type locality: Tondano, Celebes). 
Technomyrmex albipes var. vitiensis Mann, Bull. Mus. Comp. Zool., Harv. 64 : 473, worker 

(Type locality: Nadarivatu, Viti Levu, Fiji). New Synonymy. 
Technomyrmex albipes st. rufescens Santschi, 1928, Rev. Suisse Zool. 35: 70, fig. 1, worker 

(Type locality: Aiwa, Lau, Fiji). New Synonymy. 

DISTRIBUTION: SAMOA: Savai'i: Asau, 170m, disturbed rain forest, nest in leaves 
in tree crotch (RWT ace. 498). Falealupo, dry rain forest, nest from leaf mold in tree 
crotch (RWT ace. 430). Gagaifoumauga, rain forest, berlesates, fern roots and moss, and 
under bark of tree, 8m from ground (GE accs. 51, 52). Ga'utavai, coastal village, nest 
with $$ under bark of coconut palm, 19.111.1962 (RWT ace. 382). Lotogo, 340m, rain 
forest, strays from tree trunks and foliage (RWT accs. 451, 459). Matega, 65m, banana 
clearing, edge of rain forest, strays from boulders and rotting logs (RWT accs. 390, 401, 
404, 409). Mt. Matafa, rain forest, strays from roadside weeds, and nest under small 
epiphyte or tree trunk (RWT accs. 532, 538). Palauli (NK, 1955). Puapua (NK, 1955). 
**Safune, 600-1300 m, rain forest (1924). Samaeulu, coastal rainforest, strays from foliage 
(RWT ace. 509). Upolu: Afiamalu, 700m, workers in rotting logs (OHS, 1940) and <y<? 
at light, 10. VI. and 5.VII.1940 (OHS and ECZ) ; strays from weeds and grasses in over- 
grown garden (RWT accs. 252, 532) ; disturbed rain forest, strays from foliage (RWT ace. 
606) ; berlesate, tree moss 8m from ground (GE ace. 40). **Apia (1925). Mafa Pass, <?, 
10.11.1956. Malololelei Road, 500m (OHS, 1940; **1925). Sinaele, 400m (OHS, 1940). 
Tanumalala, 350m, disturbed rain forest, strays from foliage (RWT ace. 606). Tapatapao 
(Lanufao Trail), 400m, sugar cane (OHS, 1940). Utumapu, rain forest, nest in dry rot- 

Wilson & Taylor: Ants of Polynesia 

Fig. 68. Technomyrmex albipes (Fr. Smith). Worker from Koror (NE), 
Palau, Micronesia. 

ting log (RWT ace. 578). **Vaea, 350m (1924). **Vailima (1924). Tutuila: Afono Trail 
(W. Side), 150m (ECZ 1940). Amanave (OHS 1940). Amouli, 250-300 m (ECZ, 1940; 
**1923). Aua-Afono (Fullaway 1930). Fagatogo, 300m, beating shrubbery (ECZ 1940). 
Fagasa Trail (D. T. Fullaway, 1930). Leone-Aule (Fullaway 1930). Mt. Pioa, beating 
shrubbery, &, 29 .VIII. 1940 (ECZ). Pago Pago (N. Side), beating dead branches, 400m 
(ECZ, 1940) **Pago Pago (1924). **Fagasa (1923). Manua: Ta'u: Luma (W. Harris, 
1937). Recorded from Samoa by Mayr (1876). TONGA: Tongatabu: Nukualofa (NK, 
1956). Haamea (NK, 1956). Vava'u : Falevai (NK, 1956). Neiafu (NK, 1956). Eua : 
Ohonua (NK, 1956). Recorded from Tonga by Mayr (1876). COOK IS.: Rarotonga: 300 
m (G. P. Wilder 1929). Avarua (W. M. Wheeler 1914). TOKELAU IS. : Fakaofo (EHB, 
1924). SOCIETY IS.: Moorea (*1906-07, 1925). *Faaroa, 300m. Baie de Cook (NK, 1955). 
Tahiti: *Tuaruru. *Fautaua Vail. (1934). *Faraura Vail., Hitiaa, 150m. *Papeari. *Arihiri, 
Pare (1934). *Lake Vaihiria (1927). *Mt. Aorai Trail, 1150m (1934). Vallee de la Mission, 
3.2km inland (1925; Cheesman & Crawley 1928). Coast-hills behind Papeete (1925; 
Cheesman & Crawley, 1928). Vaitepiha Vail. (1925 ; Cheesman & Crawley 1928). Papenoo 
Vail. (NK, 1955). Raiatea: Opoa (NK 1955). *(1925). Tahaa : Tiva (NK, 1955). *NW 
ridge of Mt. Turi, 500-700 m (1934). AUSTRAL IS.: Raivavae: E. and W. slopes of Mt. 
Muanui, 150-250 m (1934). Near Unurua (1934); S. slope of Pic Rouge (1934). Tubuai: 
Rautaro Islet (1934). Rimatara: Maraitere (1934). Rurutu: SW slope of Mt. Manureva, 
300m (1934). *TUAMOTU IS.: Timo6, north islet (1934). *GAMBIER IS.: Agakauitai 
(1934). Mangareva (1934). Akamaru (1934). Tar aval (1934). MARQUESAS: Wheeler 
(1932b, 1933d) lists many records of collections, made chiefly in the Twenties, from Hiva 
Oa, Tahuata, Fatu Hiva, Moho Tane, Nuku Hiva, Ua Huka, and Eiao. The localities 
ranged from near sea level to 800 m. On the basis of comparative collections it is clear 
that T. albipes is a dominant ant species in this archipelago. *HENDERSON I. : N. and 
NW side (1934). *PITCAIRN I.: N. side (1934). *HAWAII : Wheeler (1934a) records 
this species from Kauai, Oahu, and Molokai. Curiously, it was not present in the collec- 
tions studied by Forel in 1899; it might well have been introduced through human com- 
merce in the intervening decades. 

Pac. Ins. Mon. 14 

Donisthorpe (1932: 465) identified Formica albipes Fr. Smith 1861 as synonymous with 
F. detorquens Walker 1859 and in subsequent publications (1932-1950) used the latter name. 
This is unfortunate, since albipes had been used countless times in reference to this eco- 
logically very important species, and a clear case for conserving it was evident. Other 
authors since 1932 have continued to use albipes. Now it appears that Donisthorpe's syn- 
onymy was not justified. Brown (1958) states: "In 1950, the late Mr. John Clark told 
me that he had learned that the type of Walker's detorquens was a badly damaged female 
specimen, the identity of which could not certainly be made out." In order to promote 
nomenclatural stability in a case where it is badly needed, it is here proposed to treat 
detorquens as a nomen dubium and to follow the century of continuous usage of the name 

The species is the most widespread of all the Indo-Australian Technomyrmex, ranging as 
a dominant ant from India to eastern Australia and throughout the Pacific, including Mela- 
nesia and Micronesia. 

Anoplolepis longipes (Jerdon) 

Fig. 69. 

Formica longipes Jerdon, 1851, Madras J. Lit. Sci. 17: 122, worker. 

DISTRIBUTION : SAMOA : Savai'i : Alagaogao, 100 m, plantation, tending aphids on 
cocoa flower stalks (RWT ace. 415). **Fagamalo (1925). Falealupo, dry forest, nest under 
stone (RWT ace. 428) and strays from foliage (RWT ace. 425). Fanga (NK, 1955). Ga'- 
utavai, coastal coconut plantation, nest under bark of coconut trunk (RWT ace. 383). 
Manase, native village, strays from roadside weeds (RWT ace. 507). Patamea (NK, 1955). 
**Safune (1925). Salelologa, strays around house (RWT ace. 547). Samalaeulu (NK, 1955); 
coastal rain forest, strays from vegetation (RWT ace. 509). Sili (NK, 1955). Vaisala, 
native village, strays from cocoa-fermenting bin (RWT ace. 467). Upolu: Apia (ECZ, 
1940; ** 1923). Falepuga, seashore (OHS, 1940). Lauli'i (TEW, 1956). Malololelei Road, 
400 m (OHS, 1940). **Mulifanua (1925). Sa'agafou, lowland rain forest, strays from foliage 
(RWT ace. 600). **Siumu (1923). Tapatapao, 300m (OHS, 1940). Tutuila : Afono Trail, 
W. side, 80m (OHS, 1940). Amouli, 250-300 m (ECZ, 1940; **1923). Aua (EHB, 1935 ; 
OHS, 1940). Breaker Point (OHS, 1940). Fagatogo (OHS, 1940). Reservoir, Fangatanga 
(ECZ, 1938). **Leone Road (1925). Mt. Pioa, 300m (OHS, 1940). Naval Station, male 
at light, 11.VIII.1940 (OHS, ECZ). Pago Pago (EHB, 1924; NK, 1941; **1925). Manua: 
Ta'u (A. F. Judd 1926; **1923, 1926). Recorded from Samoa by Gustav Mayr (1870). 
TONGA: Vava'u : Neiafu (NK, 1956). Recorded by Mayr (1870) from Tonga, no further 
locality. WALLIS : Recorded by Emery (1914a) from material in the Paris Museum. 
ROTUMA: Paho, 30-200 m. Saluaka, 80m. Solkope, 0-130 m. (Rotuma collections all by 
H. St. John, VIII. 1938). ELLICE IS.: Recorded by Mayr (1870). LINE IS.: Fanning (S. 
C. Ball, 1924). TOKELAU ISLANDS: Fakaofo (EHB 1934). COOK IS.: Rarotonga (C. 
E. Clarke 1937). SOCIETY IS.: Moorea: Baie de Cook (NK, 1955). *Tehau Pt. (1934). 
Tahiti: *Arihiri, Pare (1934). *Arue (1941). *Blue Lagoon (1934). *Fautaua Valley (1934). 
*Papeete (1926, 1932). Taharaa, near Pt. Venus (RWT, 1960). *Tehau Pt. (1934). Raiatea: 
*Faaroa Bay (1934). Opoa (NK, 1955). Huahine: Haapu (NK, 1955). Abundant on Tahiti 
and Raiatea in 1925 (Cheesman & Crawley 1928). * AUSTRAL IS.: Rimatara: Oromana 

Wilson & Taylor : Ants of Polynesia 

Fig. 69. Anoplolepis longipes (Jerdon), worker from Sorol I., Sorol Atoll, 
W. Caroline Is. 

Hills (1934). *TUAMOTU IS.: Tukuhora : Amaa (1934). South Marutea : NW islet (1934). 
Makatea (1934). Abundant on Fakarava in 1925 (Cheesman & Crawley 1928). *GAMBIER 
IS.: Agakauitai (1934). Mangareva: Near convent (1934). ""MARQUESAS : Numerous lo- 
calities on Hiva Oa, Tahuata, Fatu Hiva, Nuku Hiva, Moho Tane, Ua Huka, and Eiao ; 
all collections in 1920's determined by W. M. Wheeler. Abundant on Hiva Oa and Fatu 
Hiva in 1925 according to Cheesman and Crawley (1928). HAWAII: Oahu : Barber's 
Point, queen and workers (P. W. Weber, XII.1952). Zimmerman (1953) identified workers 
collected at the same locality by E. J. Ford, Jr., from Hibiscus on 17.IV.1952. Since the 
species is so distinctive in appearance and Hawaii has been generally well collected, there 
is little doubt that the Barber's Point population represents a relatively young (or ecolog- 
ically very restricted) infestation. 

A. longipes is native to Africa and has been spread by human commerce throughout most 
of the Old World tropics. It is a dominant ant in disturbed habitats in Melanesia and 

Plagiolepis alluaudi Forel Fig. 70. 

Plagiolepis alluaudi Emery, 1894, Ann. Ent. Soc. France 63: 71, worker (Type locality: La 

Misere, Make, Seychelles). Smith, 1957, J. N. Y. Ent. Soc. 65: 195-198, taxonomy, 

Plagiolepis mactavishi Wheeler, 1908, Bull. Amer. Mus. Nat. Hist. 24 : 166, worker (Type 

locality: Moorea). 
Plagiolepis for -eli Santschi, 1920, Bull. Soc. Vaud. Sci. Nat. 53: 165, figs. G, H, I, worker 

(Type locality: Botanical Garden, Zurich, Switzerland). 
Plagiolepis foreli var. ornata Santschi, 1920, Bull. Soc. Vaud. Sci. Nat. 53 : 166, worker 

(Type locality: New Caledonia). 
Plagiolepis augusti Emery, 1921, Ann. Ent. Soc. Belg. 61 : 317, Norn. nov. pro foreli Mann. 

Cheesman & Crawley, 1928, Ann. Mag. Nat. Hist, ser, 10, 2 : 524. 

Pac. Ins. Mon. 

Plagiolepis foreli : Mann. 1921, Bull. Mus. Comp. Zool., Harv. 64: 473, worker (Type lo- 
cality: Suva, Viti Levu, Fiji). 

DISTRIBUTION: *SOCIETY : Moorea (1906-07). Tahiti: Valley of Pirae, Pare (1934). 
Raiatea: (1925). Uturoa (1925). Huahine (1925). Cheesman & Crawely (1928) record this 

species (as augusti Emery) from collections 
/) made in the Vallee de Ste. Amelie, Tahiti, in 

J/ 1925. * AUSTRAL IS. : Rapa : two-fifth kilo- 

/ 1 1\ meter E of Area (1934). NE ridge of Mangaoa 

, II \ \X ~^>-- Peak (1934). Raivavae: Pic Rouge, S. slope 

L \\ r\ \C Y ^~"\ / (1934). *GAMBIER IS.: Akamam: N. side 

\J } VJ \] /^^ \ P (1934). Mangareva: NE slope of Mt. Duff 

^ \ (1934). *MARQUESAS: Hiva Oa: Hanaheka, 

^ ^^ -JQ 350m. Atuona, 400m. Vaioa, 400 and 600 m 

EK ( in 1920's). Cheesman & Crawley (1928) re- 

cord the species from Tahuata in 1925. EAST- 
Fig. 70. Plagiolepis alluaudi Emery, syn- ER j . Recorded by Wheeler (1922). Collected 
type worker of synonymous P. mactavishi Mathias h f j_ 

Wheeler from Moorea, Society Is. * . 

lowing localities in I-II 1965 : in Zea mats, 

Hanga Roa ; outer slopes of Rano KSLO, litter under Eucalyptus ; in crater of Rano Kao ; 
in crater of Rano Raraku. HAWAII : *Oohu : Honolulu. Koko Head. *Molokai : forests. 
*Kauai: Kapaa, in pineapple fields. Suehiro (1960): Midway. Laysan: workers and alate 
queens, IX.1961 (G. D. Butler). 

According to M. R. Smith (1957) this pantropical tramp species is steadily increasing 
its range through transport with human commerce. It is apparently native to Africa. 
Specimens have recently been recorded from St. Helena, in the South Atlantic (Taylor & 
Wilson 1962), as well as Suva, Fiji (W. M. Mann 1914; NK 1949, 1950) and Szechwan 
Province, China (W. L. Brown 1945). 

Plagiolepis exigua Forel Fi_ 

Plagiolepis exigua Forel, 1894, J. Bombay Hist. Soc. 8: 415, worker, queen (Type locality: 

Poona, India). 

^ - -. , -/I DISTRIBUTION : *HAWAII : Oahu : Makiki. Kalihi. 

f I I \^ Aiea. Tantalus. Hawaii : Naalehu. 

1 Exigua has been recorded from India, western China, 
Hong Kong, Madagascar, and Ethiopia, as well as from 

, r\ A I / . Hawaii. It can be distinguished from the much more 

\\J , I I (J I '4 common and widespread alluaudi by its shorter antennal 

\ ~"jA/^\|;Cl"/ \ \ scapes, sparser pilosity, and shinier body surface. The 

<^/^ ~^^ \ above Hawaiian records have not been verified by us. 

N\-- r -^/ 7 1 The specimens used by Wheeler have apparently been lost, 

^T 7 ^ and no further collections from Hawaii have come to our 

Fig. 71. Plagiolepis exigua notice. It is entirely possible that Wheeler misidentified 

Forel, syntype worker from samples of P. alluaudi as this species. 
Poona, India. 

1967 Wilson & Taylor : Ants of Polynesia ! 

Paratrechina (P.) longicornis (Latreille) Fig. 72. 

Formica longicornis Latreille, 1802, Fourmis, p. 113, worker. 

DISTRIBUTION: SAMOA: Savai'i: Puapua (NK, 1955); **Safune 600-1800 m (1924). 
Upolu: Togitogiga, rain forest, berlesate, base of pencil orchid, 12. VI. 1962 (GE ace. 55). 
Tutuila: Malaeini, rain forest berlesate, moss on rotting logs, 13.VII.1962 (GE ace. T21). 
Pago Pago (1918). **Leone Road (1925). Mayr (1870) recorded the species from Samoa 
without specifying any locality. SWAINS I.: Sweeping Scaevola frutescens (ECZ, 1940). 
TONGA: Tongatabu: Nukualofa (NK, 1956). Tongatabu, no further locality (G. P. Wilder 
1923). TOKELAU IS.: Fakaofo (EHB, 1924). PHOENIX IS.: Canton (E. O. Wilson ex- 
tremely abundant around airstrip in 1955 ; NK 1958). " The dominant ant in the inhabited 
part of the island" in 1940-41 (van Zwaluwenburg 1943). DANGER IS.: Motu Kotawa 
(EHB, 1924). *Puka Puka (1933). COOK IS. : Rarotonga (G. P. Wilder 1925). LINE IS. : 

Fig. 72. Paratrechina longicornis (Latreille), worker from United States. (From 
S. H. DeBord in M. R. Smith, 1947). 

Jarvis (EHB, 1935). Palmyra: E. J. Ford (1959) records this species on sprouted coconuts 
brought from Palmyra to Honolulu in 1957. Johnston: 1947-52 {fide Chilson 1953). *SO- 
CIETY IS.: Moorea (1906-07). Tahiti: Teohu Vail, (collector unknown). 1.6km S of 
Papeete on coast road (RWT, 1960). Cheesman & Crawley (1928) report collections made 
in the Vallee de Ste. Amelie, Tahiti, and on the coast of Bora Bora in 1925). MARQUE- 
SAS : Wheeler (1932b) reports collections from Moho Tane (60-100 m), Ua Huka, Ua 
Pu, and Eiao (4500m) made in the 1920's. Cheesman & Crawley (1928) report it occur- 
ring on Tahuata in 1925. *OENO I.: (1934). HAWAII: Perkins, in Forel (1899), stated 
that in the 1890's longicornis was " abundant all over the plains." Wheeler (1934a) cites 
numerous records from Oahu, Maui, Kauai, Hawaii and Nihoa I. Suehiro (1960) records 
it more recently from Midway. 

P. longicornis is one of the most widespread and abundant of all pantropical ant species. 
It is especially well adapted to dry habitats and abounds in the most urbanized portions 
of many tropical towns and cities. It probably originated within the Old World tropics, 
perhaps specifically in southeastern Asia or Melanesia. The most closely related species 
appears to be P. (Paraparatrechina) pallida Donisthorpe of New Guinea. 

Paratrechina (Nylanderia) bourbonica (Forel) 

Prenolepis nodifera st. bourbonica Forel, 1886, Ann. Ent. Soc. Belg. 30: 210-11, worker, 

Pac. Ins. Mon. 

queen, ft (Type locality: St. Denis, Reunion). 
Prenolepis bourbonica r. bengalensis Forel, 1894, J. Bombay Nat. Hist. Soc. 8: 406, 407-408, 

worker, ft (Original localities: Calcutta, India; Burma). New Synonymy. 
Prenolepis bourbonica r. hawaiensis [!] Forel, 1899, Fauna Hawaiiensis, (Heterogyna : For- 

micidae), pp. 120-21, worker, queen, ft. (Original localities: Oahu, Molokai, Hawaii). 

New Synonymy. 
Prenolepis (Nylanderia) bourbonica subsp. skottsbergi Wheeler, 1923, Nat. Hist. Juan Fernandez 

3: 318, worker, ft (Type locality: Easter I.). New Synonymy. 

DISTRIBUTION : SAMOA : Savai'i: Ga'utavai, native village, nest with 3*3- and winged 
queens under bark of coconut palm, 19. III. 1962 (RWT ace. 379). Puapua (NK, 1955). 
Salelologa, strays around house (RWT ace. 547). Vaisala, strays inside house (RWT ace. 
446). Upolu: Apia (OHS, 1940); $$ and winged queens in light trap, 4-9.IV.1962 (GE). 
Utumapu, rain forest, strays on tree trunks (RWT ace. 564). Tutuila : **Amouli (1923). 
**Leone Road (1926). Naval Station, <?<? at light, 12.VIII.1940 (OHS, ECZ). **Pago 
Pago (1923). Tunulu, seashore (OHS, ECZ, 1940). Manua : Olosega (D. T. Fullaway 
1930). TONGA: Tongatabu: Nukualofa (NK, 1955). PHOENIX IS.: Canton, 1940-41, 
fide van Zwaluwenburg (1943). Hull (EHB, 1924). COOK IS.: Aitutaki: Akaiami (NK, 
1955). Rarotonga (G. P. Wilder, 1925). LINE IS.: Fanning: workers and a single ft, 
VII. 1922 (S. C. Ball 1922). SOCIETY IS.: Tahiti: Papenoo Vail. (NK, 1955). Pt. Venus 
(C. H. Edmondson, 1933). 1.6km S of Papeete on coast road (RWT, 1960). *Tuauru Val. 
*Blue Lagoon, near Papeete (1934). Raiatea: Uturoa (NK, 1955). Huahine: Haapu (NK, 
1955). Cheesman collected bourbonica near Papeete, in the Vaitepiha Valley, Tahiti, on 
Raiatea, on Motu Moute (near Bora Bora), and below Mt. Papoti on Bora Bora, all in 
1925 (Cheesman & Crawley, 1928). AUSTRAL IS.: Raivavae, in house (A. M. Stokes, 
1922). *Tubuai: Murivani (1934). *TUAMOTU IS.: Tekotiko: Hao (1934). Tepoto : 
near center (1934). *GAMBIER IS.: Taravai (1934). Mangareva : near convent, 100m 
(1934). Agakauitai: (1934). MARQUESAS: Numerous records from Tahuata, Fatu Hiva, 
and Ua Huka ; workers at sea level, \ft captured at light at 900 m (Wheeler 1932b, 
1933d). Collected on Tahuata, Nuku Hiva, and Fatu Hiva in 1925 (Cheesman & Crawley 
1928). *PITCAIRNI. : S. side, 200^300 m (1934). *FLINT L: (1934). EASTER: (*1922; 
see comment below). I. E. Efford and J. A. Mathias collected the species at the follow- 
ing places in I-II.1965 : Hanga Roa, 4 collections ; garden of Carlos Rapu ; on grass, 
hillside behind beach at Ovahe ; Motu Nui ; Rano Raraku. Winged $& were found 
on 19, 29, and 30.1. and l.II. A winged queen was found 22.1. HAWAII: "Oahu, 
Molokai, and Hawaii; from the coast to 400ft." (Perkins, in Forel, 1899). Numerous 
records from Oahu, Hawaii, Maui, Kauai, Midway, and Wake (*1934a). 

The male of bourbonica is easily recognized by its distinctive genitalia : the parameres 
are broad, short, and with excised outer margins, while the volsellae are unusually heavy 
and broad, and darkly colored. The workers are consistently large, with long scapes that 
surpass the occipital corners by almost exactly half their own length, and concolorous 
dark brown. Most workers are most readily separated (in the absence of males) from 
the similar P. vaga by their size ; the head width of the great majority of individuals range 
between 0.65 and 0.72mm. Vaga males have relatively longer, tapered parameres with 
entire outer margins and thinner, more lightly colored volsellae. The workers of vaga are 
extremely variable in the characters just cited with reference to bourbonica ; in the study 

Wilson & Taylor : Ants of Polynesia 

series a few workers can be found that overlap bourbonica in individual characters. The 
great majority of vaga workers have head widths between 0.45 and 0.62mm and slightly 
shorter scapes than bourbonica, and are light to medium reddish brown in body color. 
It is to be admitted, however, that a few individuals (e. g., " vaga " from Fanning) 
cannot be placed with certainty in either species. 

Bourbonica apparently originated from tropical Asia and has been spread by commerce 
throughout the Indian and Pacific Oceans and to a few localities in the New World tropics. 
On Samoa it favors more disturbed habitats than either minutula and vaga. 

Paratrechina (Nylanderia) minutula (Forel) Fig. 73. 

Prenolepis minutula Forel, 1901, Mitt. Zool. Mus. Berl. 2: 25, nota, worker (Type locality: 

New South Wales, Australia). 

Prenolepis minutula r. atomus Forel, 1901, loc. cit., worker (Type locality : Ralum, New 
Britain). New Synonymy (provisional). 

DISTRIBUTION : SAMOA : Upolu : Utumapu, rain forest, nests with &&, ex rotting 
wood fragment and under epiphyte on trunk of sapling, 18. III. 1962 (RWT ace. 575). 
Vaipoto, rain forest, leaf litter and 

soil (TEW, 1956). Tutuila: N of /J 

Aloau, 450m, stunted forest, workers ^ ' ' *~ 

and 5^, 13.VII.1962 (GE aces. T16, / // // ,f ,/ , 

T19). Fagatogo, 300 m, beating / /A J^ 

shrubbery (ECZ, 1940). Fangatanga, Lq /C\ / \ ' X 

Reservoir (ECZ, 1938). Le Mafa, \J \ /// \ /' 

250m, rain forest, berlesates, epi- \ I A u/ 
phytes, leaf litter, moss on newly \ < ^ 
fallen and rotting logs, 12.VII.1962 ^^ 
(GE aces. T1-T8, T10). Malaeini, ( 

rain forest, berlesates, moss on rot- 3 _ ParatreMnaminutula(fmer) , Worker from 

ting logs, 13.VII.1962 (GE aces. T21, Uteleif Tutuila (ECZ) . 
T24). Naval Station, males at light, 

11-15. VIII. 1940 (ECZ). **Pago Pago (1925). Between Oloava and Olomanu, 430 m, rain 
forest, moss on rotting log, litter, and soil, 13. VII. 1962 (GE aces. T12, T14). Utulei, 150- 
200m (ECZ 1940). 

Among Indo-Australian series of the minutula complex in the Museum of Comparative 
Zoology, the only character showing variation of possible species significance is in size. 
Workers from Rottnest I., Western Australia and Lord Howe I. (" typical " minutula) are 
larger, with head width ranging 0.39-0.40 mm. Series from Brisbane, Queensland ; Lampong 
Districts, southern Sumatra ; Ugi, Solomon Is. ; and Samoa (form " atomus ") have head 
widths ranging 0.33-0.38 mm. This difference does not seem sufficient to warrant specific 
separation at the present time. Among the smaller (" atomus ") variant, color varies from 
medium brown (Queensland, Samoa) to yellowish brown (others). The geographic origin 
of minutula is unknown, although the species is probably native to some part of the great 
Indo-Australian area. 

Paratrechina (Nylanderia) sharp! (Forel) 
(Doubtful establishment) 

Pac. Ins. Mon. 

Prenolepis sharpii Forel, 1899, Fauna Hawaiiensis (Heterogyna : Formicidae), p. 121, figs. 
1-1 a, worker, queen, # (Type locality: Honolulu). 

DISTRIBUTION : HAWAII : Oahu: "Honolulu, brought with plants from China " (Forel 

The species is very close to vaga, evidently being distinguishable only in the male caste. 
Judging from Forel's figure, the parameres are distinctively narrower and more tapered 
than in vaga, and the ventral margins are crenulated. Since the type series came from 
what was evidently a single colony introduced from China more than 60 years ago, the 
establishment of sharpi in Hawaii must be considered in great doubt until additional 
material is collected. 

Paratrechina (Nylanderia) stigmatica (Mann) 
(Doubtful record) 

Prenolepis (Nylanderia) stigmaticus Mann, 1919, Bull. Mus. Comp. Zool., Harv. 63: 367-68 
(Type locality: Wai'ai, San Cristoval, Solomon Is.) 

DISTRIBUTION: SAMOA: "Upolu: Vailutai, 1 incomplete worker, 28.IV.1924 (Bryan)" 
(Santschi 1928). 

This single record must be considered very doubtful, especially since the much more 
extensive ant collections made in 1938-40 and 1956-62 contain no further series. Stigmatica 
is a distinctive species, the worker being marked by a flattened mesothorax and unusually 
long, slender scapes, which surpass the occipital corners by 5/8 their length. It is still 
known with certainty only from the Solomon Is. 

Paratrechina (Nylanderia) vaga (Forel) Fig. 74. 

Prenolepis obscura st. vaga Forel, 1901, Mitt. Zool. Mus. Berl. 2: 26, worker (Type locality: 

Ralum, New Britain). 

Prenolepis vividula : Mayr, 1876, J. Mus. Godeff, Hamburg 12: 23, distribution in Pacific 
(Nee Formica vividula Nylander). Mann, 1921, Bull. Mus. Comp. Zool., Harv. 64 : 476 

(Distribution in Fiji, Nee Formica vividula 
yf\ Nylander) . 

Paratrechina (Nylanderia') vaga var. crassipilis 

{ H, l\ \\ Santschi, 1928, Rev. Suisse Zool. 35: 71, 

%_. worker, 1928, Insects of Samoa, 5: 54-55, 

:T\ J ^^ fi s - 7 b-c, 8d-e, worker, dist. in Samoa 

_/ / - - ;. //^ (Original localities : Lau (many localities) 

^ - " ^ ' >j^ and Suva Bay, Viti Levu, Fiji). New 

III^- '^^ Synonymy. 

Fig. 74. Paratrechina vaga (Forel). Based Paratrechina (Nylanderia) vaga var. irritans 
on workers from Tonga. Santschi, 1928, Insects of Samoa 5: 54-55, 

figs. 7e, 8a-c (Original localities : Savai'i, 
Tutuila, Upolu, and Manua, Samoa). New Synonymy. 
Paratrechina (Nylanderia) sharpi: Wheeler, 1934, B. P. Bishop Mus., Occ. Pap. 10: 18 (1 Nee 

Prenolepis sharpii Forel. 

(?) Paratrechina vitiensis: Cheesman & Crawley, 1928, Ann. Mag. Nat. Hist. ser. 10, 2: 
525 (Nee Paratrechina vitiensis Mann). 

Wilson & Taylor : Ants of Polynesia 

DISTRIBUTION: **SIKAIANA (1933). SAMOA: Savai'i: Asau, 170m, disturbed rain 
forest, nest in rotting log (RWT ace. 490) ; strays from foliage and forest floor (RWT 
aces. 487, 501). Fanga (NK, 1955). Falealupo, dry rain forest, nests in rotting logs, small 
wood fragments and under stone, same with 3-3- and/or alate $$, 20.111.1962 (RWT aces. 
429, 442, 444, 445) ; strays from foliage and forest floor (RWT aces. 424, 427, 435, 440). 
Gagaifoumauga, rain forest, berlesates, fern roots from rocks and trees, moss from rotting 
log (GE aces. SI, S2). Lotogo, 340m, rain forest, nests from rotting logs and under stone 
(RWT aces. 465, 482, 484) ; strays from rotting log (RWT aces. 455, 456, 483) ; strays from 
foliage (RWT ace. 451). Matega, 65m, new banana clearing, edge of rain forest, 20.111. 
1962, strays from foliage and on ground, rocks, and rotting logs (RWT aces. 398, 401, 402, 
409) ; several claustral $ under moss or leaf mold on boulders (RWT ace. 391). Mt. 
Matafa, 700m, rain forest, nests from rotting log and under small epiphyte on tree trunk 
(RWT aces. 521, 539) ; strays from foliage (RWT ace. 522); berlesate of leaf mold (RWT 
ace. 520). Mt. Matafa Road, 200m, rain forest, 19.111.1962, nests with males under moss 
on stump and in rotting log (RWT aces. 368, 373) ; strays from logs and foliage (RWT 
aces. 361, 365, 374). Mt. Olomanu, rain forest, berlesate, moss and epiphytes from fallen 
tree (GE ace. S10). Patamea (NK, 1955). Puapua (NK, 1955). **Safune, low forest 
(1924). **Salailua (1924). Samalaeulu, coastal rain forest, nests and strays under small 
epiphytes on tree trunks (RWT aces. 509, 512, 515, 516) ; berlesate of leaf mold and soil 
(RWT ace. 596). Sili (NK 1955). Upolu : Afiamalu, 700m, sweeping, workers and queens, 
11. VI. 1940 (OHS, ECZ) ; leaf mold berlesate (TEW, 1956); **workers 1924; berlesates, 
ground moss (RWT aces. 2317, 2318, 2321 ; GE aces. 46, 48, 51, 53) ; berlesates, tree moss 
2-3 m from ground (RWT ace. 262) 10-12 m from ground (RWT aces. 2306, 2308; GE 
ace. 37). This was the ant most frequently encountered by Taylor at Afiamalu; it was 
collected in all major habitats visited within slightly to extremely disturbed rain forest, 
in overgrown European gardens, and in native Taro patches and a banana plantation. 
Nests were taken in rotting stumps or logs and in small wood fragments on the ground, 
as well as under moss cushions on logs and tree trunks (RWT aces. 239, 240, 248-250, 
271, 305, 307, 312, 324, 326, 330, 338). Alate 33 and queens were taken from several 
colonies collected between 9 and 17.111. **Apia (1924). Fagaloa, 170m, rain forest, ber- 
lesate of moss from logs (GE ace. 67). Falefa (NK, 1955). Falepuna (TEW, 1956). 
Lauli'i (TEW, 1956) Malololelei (NK, 1955), (TEW, 1956), **1925. Matautu (TEW, 1956). 
Poutasi (TEW, 1956). Sa'agafou, lowland rain forest, strays from foliage (RWT aces. 610, 
611). SaleaPuga (TEW, 1956). Tanumalala (TEW, 1956). Tapueleele (NK.1955). Tiapapata, 
berlesates of moss on logs and trees (GE aces. 69, 70-72, 75, 76). Togitogiga, 400m, ber- 
lesates, base of pencil orchid, leaf mold, and moss on rotting logs (GE aces. 55-60). 
Utumapu (TEW, 1956) ; rain forest, nests and strays under epiphytes on tree trunks (RWT 
aces. 566, 571, 579). Vaipoto (TEW, 1956). Tutuila: N of Aloa, rain forest, berlesates of 
moss from logs (GE aces. T17, T18, T20). Amanere, 100m, beating (ECZ, 1940). Amouli 
(D. T. Fullaway 1930). **1923. Aua-Afono (Fullaway, 1930). Breaker Point (OHS, 1940). 
Fagaitua (EHB, 1935). Faga'togo, 300m (OHS, 1940). Le Mafa, 250m, rain forest, ber- 
lesate of soil and leaves (GE ace. T7). Maupasaga-Fagasa trail (Fullaway, 1930). Mt. 
Pioa, 300m (OHS, 1940). Naval Station, # at light, 19.VIII.1940 (OHS). Between Oloava 
and Olomanu, 430m, berlesates of epiphytes and moss from trees (GE aces. Til, T15). 
Pago Pago (EHB 1924; NK 1941; **1923). Pago-Matafao trail, 400m, beating shrubbery 
(ECZ, 1940). Tafuna (TEW, 1956). Manua: Ta'u (NK, 1955; **1923). Vaga was recorded 

y/ Pac. Ins. Mon. 14 

from Samoa by Mayr (1876) as vividula Nylander. The T. E. Woodward material was 
collected mostly from berlesates of soil and leaf litter in rain forest ; one record (Falepuna) 
is from the edge of a swamp. SWAINS I.: (EHB, 1935). TONGA: Tongatabu: Nuku- 
alofa, 3tf, 11.1956 (NK). Vava'u: Neiafu NK, 1956). Pangai (NK, 1956). Eua: Ohonua 
(NK, 1956). Recorded from Tonga by Mayr (1876). ELLICE IS.: Recorded by Mayr 
(1876). COOK IS.: Rarotonga (1937). TOKELAU IS.: Fakaofo (EHB 1924) PHOENIX 
IS.: Hull (EHB, 1923). DANGER IS.: Motu Kotawa (EHB, 1924). LINE IS.: Fanning 
(S. C. Ball, 1924). Johnston: What is probably this species was recorded from Johnston 
by M. R. Smith as " Paratrechina (Ny lander ia) sp." (Chilson 1953). SOCIETY IS *Huahine- 
NW ridge of Mt. Turi, 500-600 m (1934). *Mehetia: Fatia-po, 150-300 m (1934). *Moorea 
(1906-07; 1925). Tahiti: *Arihiri, Pare (1934). *Faraura Valley. *Maara Papiera *Paea' 
Papenoo Valley. Papeete (ECZ, 1934; *1920). *Valley E of Pirae, Pare (1934) Rimatu 
(1925). *Tmpi Bay, Papeari (1934). *Tuauru. *Vallee de la Reine. Cheesman & Crawley 
(1928) erroneously list this species as vitiensis Mann, a distinct Fijian species. Collections 
of vaga were made by Cheesman in 1925 near Papeete, at 1000m, and Vaitepiha Valley 
Tahiti, and on Raiatea at about 650 m. *AUSTRAL IS. : Numerous records by Wheeler 
(1936) from Rimatara, Rapa, Maria I. (NE islet), and Raivavae, all based on collections 
made by Zimmerman in 1934. The relative frequencies of records suggest that the species 
is especially abundant on Rapa. *TUAMOTU IS.: South Marutea, NW islet (1934) Timo& 
N. islet (1934). *GAMBIER IS. : Taravai (1934). Aukena: Korovao (1934) Mangareva 
(1934). ^MARQUESAS: Wheeler (1932b, 1933d) lists numerous collections made in 
the 1920's from Hiva Oa, Tahuata, Fatu Hiva, Nuku Hiva, Ua Huka, and Ua Pu. Ele- 
vations of worker collections on these islands range from sea level to 1000m HENDER 
SON I.: NW side (1934). *FLINT I.: (1934). HAWAII: Oahu: Honolulu (Lewis 1910) 
Manoa Vail., Honolulu (1910). 

Vaga ranges more or less continuously from Queensland, the Philippines, and New Guinea 
across the Pacific to Juan Fernandez. It shows strong internidal variation in total size, 
convexity of thoracic dorsum, pilosity, density of cuticular shagreening, and depth of color' 
As noted in an earlier study of the Rennell, Solomon Is., population (Wilson, 1960), the 
variation is nearly continuous, embracing none of the forms identical with Nylanderia spe- 
cies known to be sympatric with vaga elsewhere. But the genus is notoriously difficult 
taxonomically, and all of the Indo-Australian species need to be carefully studied to ascer- 
tain the exact status and origin of vaga. In particular, the Australian obscura (Mayr) and 

Brachymyrmex obscurior Forel, n. status Fig. 75. 

Brachymyrmex heeri var. obscurior Forel, 1893, Trans. Roy. Ent. Soc. Lond. p. 345, worker 

(Type locality: St. Vincent, West Indies). 
Brachymyrmex heeri var. aphidicola, Wheeler, 1934, B. P. Bishop Mus., Occ. Pap. 10: 17 

(? nee aphidicola Forel). 

DISTRIBUTION: SAMOA: Savai'i: Puapua (NK, 1955). Upolu: Apia, workers and 
<?(?, 11.1955 (NK): workers only (ECZ, 1940) ; tftf at light, 4-9.IV.1962 (GE). Le Mafa 
disturbed roadside forest, under moss on tree trunk, 30.111.1962 (RWT ace. 635). Tapatapao 
(OHS, 1940). HAWAII: Oahu: Honolulu, in orchid baskets (H. Lyon 1914). 

The material cited above belongs to the form generally referred to in the literature as 

Wilson & Taylor : Ants of Polynesia 

obscurior. Its recognition as a distinct species here is purely a provisional measure, con- 
tingent upon a fuller revision of the large and 

difficult genus to which it belongs. For the x/'H\ >^r~~v^ /^~~^^ 

moment it can safely be said that obscurior is ~J/ DK \/ ^\ I / /^ 

native to the New World tropics and probably JKJ S~) ^\^~^^ J^^^-_/ / A ;; 

specifically to Central America and Tropical -A.^^.^/^^ ~~^\LF~rf!^ 

Mexico. It is one of the most widespread ants / tif^^ 75 

in the West Indies and has been introduced into Fig. 75. Brachymyrmex obscurior Forel. 

Florida (M. R. Smith 1951) and the port of Mo- Based on workers from Tapatapao, Upolu 

bile, Alabama (records from 1949, E. O. Wilson). ( HS )- 

The Samoan records cited above are the first for Brachymyrmex from the South Pacific. 

Camponotus (Myrmoturba) chloroticus Emery Fig. 76. 

Camponotus maculatus subsp. chloroticus Emery, 1897, Ann. Mus. Civ. Stor. Nat. Geneva 38: 

574, worker (Type locality: Irupara, New Guinea). 

Camponotus (Myrmoturba} maculatus chlorotica var. chlorogaster Emery, 1914b, Nova Cale- 
donia 1(4) : 429, worker, soldier (Original localities : Espiritu Santo and Ambrim, New 

Hebrides). New synonymy. 
Camponotus (Myrmoturba} maculatus pallidus var. Samoensis Santschi, 1919, Bull. Soc. Vaud. 

Sci. Nat. 52: 326-27, worker, soldier (Type locality: Apia, Upolu). New Synonymy. 
Camponotus (Myrmoturba} maculatus subsp. sanctae crucis Mann, 1919, Bull. Mus. Comp. 

Zool., Harv. 63 : 369, fig. 42, soldier (Type locality : Santa Anna, Graciosa Bay, Santa 

Cruz I.). New Synonymy. 
Camponotus novaehollandiae, Mayr, 1870, J. Mus. Godeffroy, Hamburg 12: 66. INec novae- 

hollandiae Mayr, 1870, Cape York, Queensland. 

DISTRIBUTION : SAMOA : Savai'i: Falealupo, rain forest, nest with alate queens under 
bark of rotting log, 20.111.1962 (RWT ace. 431). Lotogo, 350 m, rain forest, nest in hollow 
twig on tree and strays on tree trunk, 21.111.1962 (RWT aces. 470, 471). Mt. Matafa, 
700 m, rain forest, nest in dry, hollow branch on tree, 23.111.1962 (RWT ace. 525). Pa- 
lauli (NK 1955). Patamea (NK 1955). **Safune (1925). Sili (NK 1955). Taga, coastal 
Futu grove, nest under epiphyte 1.5 m above ground, 19.111.1962 (RWT and GE ace. 350). 
**Tuasivi (1924). Upolu : Afiamalu, 700m, alate queen on grass in evening, 10.111.1962 
(RWT ace. 261). **Apia (1924, 1925). Falefa (ECZ 1940). Le Mafa, roadside, nest 
under moss on tree trunk 2 m above ground, 29.111.1962 (RWT ace. 635). Mo'ata, near 
Apia, colonies in twigs of mangrove, 18.111.1962 (RWT aces. 548, 550, 551, 553, 556). Ma- 
lololelei Road, 500 m, winged queens (OHS, ECZ, 1940). **Mulifamua, # (1925). Sa'- 
agafou, lowland rain forest, nest in twig on tree, 28.111.1962 (RWT ace. 610). Tapatapao, 
300m, dead log (OHS 1940). Tapueleele (NK, 1955). Tutuila: Afono Trail, W. Side, 
150 m (ECZ, 1940). **Amouli (1923, 1926). Amanave, 130 m (OHS, 1940). **Fagasa 
(1923). Fagatogo, 300 m, beating dead branches (ECZ, 1940). Naval Station, 3-3- at 
light, 12.VIII.1940 (OHS, ECZ). Pago Pago (P. J. Darlington 1931; EHB 1935; **1923). 
N. Side, Pago Pago, 200 m, on tree fern (ECZ, 1940). Mt. Pioa, 300 m (OHS, 1940). 
Manua Ta'u: Luma, <?<? at light, 11.IX.1937 (W. Harris). **Ofu (1926). The species is 
recorded by Mayr (1870) as novaehollandiae from Samoa, with no further locality cited. 
TONGA: Tongatahu: Nukualofa, workers and <?<?, 11.1956 (NK). Vava'u: Holonga (NK 

Pac. Ins. Mon. 

1956). No locality (G. P. Wilder 1925). Eua : Ohonua (NK 1956). Haapai: (G. P. 
Wilder 1925). Recorded by Mayr (1870) as novaehollandiae from Tonga, with no specifi 

island cited. ELLICE IS. : Re- 

\ corded by Mayr (1870) as novaeho- 

~~~^ (i/ llandiae. ROTUMA : Fapufa (H. 

{\ j St - John 1938 )- "Kilinga" (=Kil- 

X S 1J g a? ). 60 m (St. John 1938). Pa- 

f _ f) ho, 30-200 m (St. John 1938). 

I ~ll\ S6lk P e> sea level to 13 m ( St 

\ x , / // \ John 1938). DANGER IS. : Mo- 


~~> ( : ]/ ^7 This l ar e ' conspicuous species 

y [ I ^^^ ~~~ \ Q\ I M L ranges from New Guinea through 

\ / - \ / Micronesia and outer Melanesia 
\ S\ V / (excepting New Caledonia) to 

,,, .[ Tonga and the Danger Is. Un- 

/ fortunately, the taxonomic status 

of the population here called 

Fig. 76. Camponotus chloroticus Emery. Solkope.Rotuma chloroticus, and hence our under- 
(H. St. John) : above and left, minor worker ; right, standing of its precise distribution 
soldlen in the remainder of the Indo- 

Australian region, must remain 

clouded until the difficult maculatus group is better worked out as a whole. Suffice it to 
say that there is a species, to which the name chloroticus evidently applies, that can be dis- 
tinguished as a widespread population distinct from several related sympatric species in 
the western Pacific, including New Guinea. Whether chloroticus is in fact the oldest name 
applicable to the species cannot be stated with certainty at the present time. 

Camponotus (Myrmoturba) navigator Wilson and Taylor, new species Fig. 77. 

DIAGNOSIS AND RELATIONSHIPS : A member of the irritans group, closest to the sympa- 
tric chloroticus but differing from that species as follows : antennae proportionately longer ; 
entire body distinctly less pilose ; petiole distinctly thinner in side view ; body color pre- 
dominantly blackish brown (brownish yellow in chloroticus). Navigator also resembles papua 
Emery of New Guinea, being more similar to it than to chloroticus in pilosity and relative 
scape length ; but differing in petiole form and color, and, in addition, lacking the abrupt 
mesonotal rise just posterior to the promesonotal suture that distinguishes papua from chlo- 

Holotype major worker: HW 2.25 mm, HL 2.43 mm, CI 93, SL 1.96 mm, SI 87, PW 1.31 
mm. Characteristics as indicated in diagnosis and shown in fig. 77. The color needs 
further description : head, except mandibles, uniformly blackish brown ; mandibles deep 
"mahogany" reddish brown ; remainder of body very dark reddish brown to blackish 
brown, predominantly the latter ; scape blackish brown, remainder of appendages medium 
to dark reddish brown. 

Worker paratype variation : Strongly and continuously polymorphic, HW ranging in holo- 
type nest series from 1.00-2.25 mm. Pilosity variable: in callows many of the long hairs 

Wilson & Taylor : Ants of Polynesia 

Fig. 77. Catnponotus navigator Wilson and Taylor, holotype soldier. 

have dried in a sinuous form but in fully developed workers they are straight or smoothly 
and highly curved ; number of long propodeal hairs 1-8 in the several nest series, with 
most workers having 3-6. Body form, scape index, and coloration relatively uniform. 

Queen: HW 1.88 mm, HL 2.18 mm, SL 1.95 mm. Sharing the diagnostic worker char- 
acteristics just described, otherwise a typical queen for this species group showing the 
usual caste characteristics. 

MATERIAL EXAMINED : SAMOA : Savai'i : Matega, W. coast, 70 m, all collections on 20. 
III.1962 by Taylor & Ettershank. Ace. no. 387 : holotype worker, 7 paratype workers, dea- 
late queen. Ace. nos. 388, 389, 404, 408, 411, 412: 30 paratype workers, belonging to all 
worker subcastes. The species is named in apposition after the Navigator Is., an old Eu- 
ropean name for Samoa. 

This species is known only from Matega, an area well away from the inter-island ports 
of Savai'i. The several nest series were taken from nests under the bark of fairly sound 
logs lying on the ground, in a recently established banana clearing, about one acre in ex- 
tent. Relatively undisturbed lowland rain forest surrounded the collection site. The ants 
were abundant, 2 or 3 colonies being found on some of the larger logs. Nests contained 
pupae and larvae at various stages of development ; no alate sexuals were observed. 

Because of its apparent closest similarity to chloroticus of all the Camponotus species 
available to us in the Museum of Comparative Zoology collections, we have tentatively 
interpreted navigator as a derivative of chloroticus or a chloroticus-navigator antecedent, 
with navigator representing the first invader of Samoa and chloroticus the second. But 
because of the considerable differences between the two species, the conclusion must be 
taken cautiously and further attempts should be made to match navigator with other ex- 
tra-Polynesian species. 

Camponotus (Tanaemyrmex) variegatus (Fr. Smith) 

Fig. 78. 

Formica variegata Fr. Smith, 1858, Cat. Hym. Brit. Mus. 6: 19, worker, queen (Original 

localities: Ceylon, Singapore). 
Camponotus maculatus mills var. hawaiiensis Forel, 1899, Fauna Hawaiiensis, p. 122, worker, 

queen, <y (Type locality: Honolulu, Hawaii). 
Camponotus variegatus hawaiiensis : Emery, 1925, Gen. Ins. 183 : 96. 

Pac. Ins. Mon. 

DISTRIBUTION : HAWAII : "In and around houses in Hawaii" (Perkins, in Forel, 
. x, ^ . 1899). Wheeler (1934a) lists records 

-^ /, /^^ from Oahu, Maui, Molokai, and Hawaii. 

*^/a / 

/ \s v x \^ /T( ^ e s ' n sl e Hawaii collection was made 

^ ^ ( \\ at 1300 m. Suehiro (1960) cites a collec- 

)l tion made on Midway in 1959. E. O. 

/ \ Wilson collected winged queens at lights 

> in Honolulu in late November 1955. 

s- i. ... .::* i"7^\ T* 16 P resent usage of the name variega- 

I T '; I .;' / >v JKS with reference to the Hawaiian sam- 

I -.?::. ...;;;'. i) P^ es ' s * n t ' le oroa d sense - The Hawaii- 

\ e i ?]'[ :: )* an population appears almost certainly to 

\^ '".' I ^^s' have been introduced from southeastern 

^^ Asia, and the name hawaiiensis will pro- 

Fig. 78. Camponotus variegatus (Fr. Smith), bably fall as a synonym of one of the 
minor worker from Koko Head, Oahu, Hawaii. f 4 , , 

species of the variegatus complex con- 
cealed under the multitude of existing subsqecific and varietal names. 

Camponotus (Colobopsis) conicus Mayr 

Fig. 80. 

Colobopsis rufifrons Mayr, 1870, Verh. Zool.-bot. Gen. Wien 20: 944-45, worker, soldier, 

queen (Par tint. Nee rufifrons Fr. Smith). 
Colobopsis conica Mayr, 1876, 3. Mus. Godeffroy, Hamburg 12: 67, worker, soldier, queen 

(based on description of rufifrons, above) (Type locality: Tonga). 

DISTRIBUTION: TONGA: Tongatabu: Nukualofa (NK 1956). Eua : Okonua, worker 
and winged ^, 11.1956 (NK). Described from Tonga by Mayr (1870) without further 
specification of locality. 

^ " 7 This endemic Tongan species is related to 

/ A \z/ 1 f' dentatus Mayr of Fiji. It differs in its all- 

-'\. ^y V) black body coloration, much lighter mesoso- 

^^-~-__ mal scu lP turin g (consisting almost entirely of 

^ ' shagreening), and in the lack of well devel- 

oped propodeal spines. The 2 species are re- 
lated to guppyi Mann of the Solomon Is. ; the 

, \ ~T> j latter species is easily distinguished by its 

/ \ ? ( -- concolorous deep yellow body coloration and 

=5 ^\^_^_k f~l I / lack of P ed i ce U ar teeth. Finally, the aber- 

y~ \ s ^ ^ K. L/ rant bryani group of Fiji appears to be deriv- 

> 8O ^--"x ed from the conicus- dentatus stock. 

Fig. 79. Camponotus fla\oimlbatus Vieh- 
meyer, minor worker from Tapueleele, 
Upolu (NK). Fig. 80. Camponotus conicus 
Mayr, minor worker from Nukualofa, 
Tongatabu, Tonga (NK). 

Camponotus (Colobopsis) flavolimbatus Vieh- 
meyer Fig. 79. 

Camponotus (Calobopsis) [I] flavolimbatus Vieh- 
meyer, 1922, Arch IV Natur. 88: 218-19, 
queen (Type locality : Samoa). 

Wilson & Taylor : Ants of Polynesia 

Camponotus (Colobopsis} buxtoni Santschi, 1928, Insects of Samoa 5 : 56-57, worker (Type 

locality: Malololelei, Upolu, Samoa). New Synonymy. 
Camponotus (Colobopsis} conithorax var. nautarum Santschi, 1919, Bull. Soc. Vaud. Sci. Nat. 

52: 327, worker (?) (Type locality: Apia, Upolu, Samoa). New Synonymy. 
Camponotus (Colobopsis) rufifrons var. leucopus: Santschi, 1928, Insects of Samoa 

queen (? Nee leucopus Emery). 

DISTRIBUTION : SAMOA: Savai'i: Asau, 170m, lowland rain forest, strays on foliage, 
22.111.1962 (RWT ace. 487). Falealupo, dry rain forest, strays on foliage, 20.111.1962 (RWT 
ace. 424). Lotogo, 370 m, rain forest, strays on foliage, nest under epiphyte on Ficus tree 
35 m above ground, 21.111.1962 (RWT aces. 451, 458). Mt. Matafa, 700 m, rain forest, 
nest in dry hollow branch on tree, strays on roadside weeds, 23.111.1962 (RWT aces. 522^ 
526). Palauli (NK, 1955). Patamea (NK, 1955). Puapua (NK, 1955). **Safune, 600- 
1300m, rain forest (1924). Samalaeulu (NK, 1955). Sili (NK, 1955). Upolu: Afiamalu, 
700 m, (OHS, ECZ, 1940) ; sweeping from roadside weeds, 10.111.1963 (RWT ace. 252) ; 
foraging on leaves of Taro patch, 16.111.1962 (RWT ace. 321) ; disturbed rain forest, strays 
on tree trunks, 12-17.111.1962 (RWT aces. 264, 341). Apia (G. P. Wilder 1925). ^Malo- 
lolelei (1924, 1925). Sa'agafou, lowland rain forest, colony in twig on bush, 28.111.1962 
(RWT ace. 612) ; strays under moss on tree trunk (RWT ace. 611). Tanumalala, 350m 
disturbed rain forest overgrown by weeds, strays on foliage, 28.111.1962 (RWT ace. 606). 
Tapatapao, 300 m, queens and 3& at light, 19.VII.1940 (OHS, ECZ). Tapueleele (NK, 1955). 
**Tuaefu (1923). **Vaea, 350m (1924). Tutuila: Fagasa Trail, 200m (OHS, 1940). Fa- 
gatoga, 300m (OHS, 1940). Naval Station, &$ at light, 12.VIII.1940 (OHS ECZ) **Pago 
Pago (1924). 

In describing conithorax var. nautarum, a junior synonym of flavolimbatus, Santschi cor- 
rectly points out that the Samoan species differs from conithorax Emery of the New Heb- 
rides in the lighter coloration of the worker appendages (light reddish or yellowish 
brown as opposed to dark ferrugineous in conithorax}. It might be added that yet a bet- 
ter character is found in the cephalic sculpturing of the soldier ; in the true conithorax the 
longitudinal carinae of the anterior portion of the head are sparser and more widely spaced. 
On the basis of present evidence, flavolimbatus should be considered a distinct but closely 
related Samoan endemic species. 

Santschi was evidently unaware of the existence of the name flavolimbatus in the litera- 
ture at the time of his 1928 description. He distinguished "buxtoni" from "nautarum" chief- 
ly on the basis of "the scale lower and longer than in conithorax [nautarum} ; the thorax 
more rounded as in loa Mann." These are two of the more variable characters in our 
sample of the population. Santschi's description and figure do not appear to depart in any 
significant way from the variation shown by these samples. Another notably variable 
character is the color of the appendages ; in particular, the legs vary from nearly clear 
yellow (as described in the flavolimbatus type queen) to medium ferrugineous. In short, 
there is at present no evidence of the existence of more than a single species in the ex- 
tensive collections of Samoan Colobopsis now available. 

Camponotus (Colobopsis) nigrifrons Mayr Fig. 81. 

Colobopsis nigrifrons Mayr, 1870, Verh. Zool.-bot. Ges. Wien 20: 943-44, queen (Type local- 
ity: Tonga). 

Pac. Ins. Mon. 

DISTRIBUTION: TONGA: Tongatabu: Haamonga (NK, 1956), small media worker, 
tentative determination, see below. Described from Tonga by Mayr (1870) without further 
locality specification. 

Forel (1914) placed this species in Myrmamblys, without explanation. Emery (1925) 

placed it in the truncatus group of Colobop- 
\ sis. Regardless of its precise phyletic re- 

/ "N V /\ lationships, it fortunately has several good 

v^_ ^-^ Wo \ .-- and simple characters that should make it 

A. / easily recognizable. The following is a 

, UI , translation of Mayr's original description: 

"Queen : length 9 mm. Shining, red, man- 

, i a dibles ferrugineous, frons, vertex and abdo- 

] | 'if men black, legs entirely yellow ; sparsely 

pilose, feebly coriaceous, mandibles longitu- 
xJ vj dinally rugulose and sparsely punctate, an- 

' \ \ 'i terior part of head weakly truncate, rough- 

ly carinate-striate in longitudinal direc- 
tion ; clypeus longitudinally feebly convex, 

Fig. 81. Camponotus nigrifrons Mayr, minor quadrangular, slightly longer than broad ; 
worker from Haamonga, Tongatabu, Tonga genae convex, not angulate ; thorax [pro- 
(NK). Fig. 82. Camponotus rotumanus Wil- deum] unarmed ; petiole with somewhat 

son and Taylor, paratype worker. v ' r ,.,,. 

mcrassate scale, subquadrate, slightly broad- 
er than high ; its angles rounded ; anterior wings 9 mm i n length, subhyaline, pterastigma 
and costa ochraceous." 

The single media worker we have examined agrees reasonably well with the description 
of the type queen. It is extremely similar to flavolimbatus medias of the same size, differ- 
ing only in coloration as follows: whereas all flavolimbatus workers from Samoa thus far 
examined have concolorous piceous brown bodies, the nigrifrons media is strikingly trico- 
lorous : the head medium ferrugineous, the mesosoma and pedicel clear light ferrugineous, 
and the gaster piceous brown. The appendages of the nigrifrons individual are light to 
medium ferrugineous, as in many flavolimbatus workers. 

Nigrifrons, if our media worker has been correctly identified, is thus distinguishable only 
on the basis of a color character, albeit a clear-cut one. Its recognition as a distinct 
species here is a conservative, provisional measure. 

Camponotus (Myrmamblys) rotumanus Wilson and Taylor, new species 

Fig. 82. 

DIAGNOSIS (MINOR WORKER) : A member of the widespread and very diverse reticulatus 
group. Most closely resembling albocinctus (Ashmead) of the Philippines, differing in the 
minor caste as follows : (1) antennal scapes proportionately much longer, exceeding oc- 
cipital corners by approximately 1/2 their total length, the scape index 122-146 (in single 
albocinctus minor examined, the scapes exceed occipital border by only 1/3 their total 
lengths, scape index 95) ; (2) tibiae medium to dark brown (light yellowish brown in al- 
bocinctus) ; (3) total size probably greater (HW 1.03-1.15 mm as opposed to 0.91 mm in 
albocinctus). This diagnosis is based on comparison of the rotumanus type series with a 
single albocinctus worker, lacking a pedicel and gaster, from Dumaguete, Philippine Is. 

1967 Wilson & Taylor : Ants of Polynesia < 

Also used for comparison of additional characters were a small series of albocinctus sol- 
diers and queens from Dumaguete, Los Baflos, and Manila. Rotumanus also resembles 
fullawayi Wheeler of Guam ; it can be distinguished (in the minor worker) by its much 
larger size, markedly lighter coloration, and sparser, more appressed body pubescence. Final- 
ly, it can be easily distinguished from reticulatus itself, the dominant and most wide- 
spread member of the group in Melanesia and Micronesia, by the much lighter coloration 
and denser pubescence and body shagreening of rotumanus. 

Holotype -worker : HW 1.09 mm, HL 1.32 mm, SL. 1.43mm, pronotal width 0.91mm. 
Body form and pilosity as shown in fig. 82. Entire body covered with moderately dense, 
completely appressed pubescence. Pilosity and pubescence pale yellow. Vertex, occiput, 
mesothorax, propodeum, tibiae, tarsi and most of pedicel and gaster dark brown with 
little evidence of reddish tinge ; posterior margins of gastric tergites pale yellow ; remainder 
of body and appendages yellowish brown. 

Paratype variation : See measurements in diagnosis, above. These specimens show little 
significant variation in the other characters cited. 

MATERIAL EXAMINED: Holotype minor worker (BISHOP 7424), ROTUMA : Solkope, sea 
level to 130m, 24.VIII.1938 ; 4 paratype minor workers, same data. Soloroa, 60m, 11. VIII. 
1938, 3 paratype minor workers. Saluaka, 29.VIII.1938, 1 paratype minor worker. "Ki- 
linga" (=Killiga?), 60m, 4.VIII.1938, 2 paratype minor workers. All collections by H. 
St. John. Holotype and paratypes in Bishop Museum, paratypes in Museum of Compara- 
tive Zoology and U. S. National Museum. 

Polyrhachis (Chariomyrma) rotumana Wilson and Taylor, new species 

Fig. 83. 

DIAGNOSIS AND RELATIONSHIPS : Most closely resembling P. rere Mann of the Solo- 
mons and Santa Cruz Is. but easily distinguished from that species by the following 
characters : (1) propodeal and lateral petiolar spines thicker ; (2) median petiolar 
spine longer and its apex acute, whereas apex is obtuse in rere ; (3) Dorsal surface of 
the lateral petiolar spines coarsely and irregularly pitted, and opaque, whereas, in rere 
the same surfaces bear only scattered punctures and are feebly shining ; (4) striae (or 
rugulae) of pronotum weaker and show ; 
very weakly defined longitudinal orientation, /"""" 
as opposed to a strong horizontal orientation LS 
in rere, while the striae (or rugulae) of ' V 
the head and mesonotum are feeble and show ' 
a weakly defined longitudinal orientation 

compared to rere; (5) anterior lateral meso- 

notal angle less strongly marked than in rere. /^ 
In other respects rotumana is very similar, If 
if not identical, to rere. \ 

Holotype worker: HW 1.22mm, HL 1.44 Fig . 83 . Polyrhachis rotumana wilson 

mm, SL 1.54 mm, PW 1.15 mm. and Taylor, holotype worker, dorsal and 

Worker paratype variation: HW 1.20-1.35 r x f r views of petiole ' Fig " 84 ' P - rere 
___ , . , Mann, worker from Anuda, Santa Cruz Is. 

mm. I he 4 or more nest series represented 

in the type collection are remarkably invariable, showing only relatively minute departures 

Pac. Ins. Mon. 

is .S 8 2 

t>o i- W O C 

O K, aq cq o 

102 Pac. Ins. Mon. 1 

from the holotype. 

MATERIAL EXAMINED : Holotype worker (BISHOP 7425), ROTUMA : Solkope, sea level to 
130 m, 24.VIII.1938 ; 3 paratype, workers, same data. Fapufa, 15.VIII.1938, 1 paratype 
worker. "Kilinga" (=Killiga?), 60m, 4.VIII.1938, 1 paratype worker. Soloroa, 60m, 11. 
VIII.1938, 5 paratype workers. All collections by H. St. John. Holotype and paratypes in 
Bishop Museum, other paratypes in Museum of Comparative Zoology and U. S. National 

This interesting species is the first Polynesian member of the huge genus Polyrhachis to 
be discovered. As noted in the diagnosis, it is closest to rere of the Solomons and Santa 
Cruz Is. To date neither rere nor any cognate species has been found in the neighboring 
Banks Is., New Hebrides, or Fiji. 


The species listed in Table 3 are known only from single doubtful records or else from 
material intercepted at quarantine. In either case, it is very unlikely that permanent pop- 
ulations are now in existence on the islands where the records were made. 

The following species were intercepted at Honolulu after they had already become suc- 
cessfully established. All records are taken from Wheeler (1934a) : Pheidole megacephala, 
Solenopsis geminata, Monomorium floricola, M. pharaonis, M. gracillimum, Tetramorium guine- 
ense, T. simillimum, Technomyrmex albipes, Tapinoma melanocephalum, Paratrechina longicornis, 
P. bourbonica. These collections are significant in that they show that the introduction of 
pantropical "tramp" species is a recurring process. 


In a first complete analysis of the ants of New Zealand, Brown (1958) showed the na- 
tive fauna to be much smaller than one would expect to find on warm-temperate islands 
of such considerable size. Moreover, the few native stocks are almost exclusively of Aus- 
tralian origin, thus differing conspicuously from the native ants inhabiting the remainder 
of western Polynesia. In addition New Zealand contains a considerable number of intro- 
duced species. Brown (1958), Cumber (1959) and Taylor (1959, 1961) have shown that 
some of these are the familiar tropical tramp species distributed elsewhere in the Pacific, 
while others have been brought in directly from Australia and are as yet unknown from 
other parts of the world. In Table 4 we have listed the 32 species recorded from New 
Zealand to date, together with an estimation of their origins. 

Table 4. List of the Ants of New Zealand. 

Myrmecia brevinoda Forel. Introduced from E. Australia. Brown (1958). 

Amblyopone australis Erichson. Probably introduced from Australia. Brown (1958). 

Amblyopone saundersi Forel. Endemic. Brown (1958). 

Heteroponera brouni Forel. Endemic. Brown (1958). 

Rhytidoponera metallica (Fr. Smith). Introduced from E. Australia. Taylor (1961). 

Discothyrea antarctica Emery. Endemic. Brown (1958). 

Mesoponera castanea (Mayr). Endemic. Brown (1958). Taylor (unpublished notes) finds that 

967 Wilson & Taylor : Ants of Polynesia 10: 

this name applies to a species with ergatoid queens, apparently limited to the North I. 
Mesoponera castaneicolor (Dalle Torre). Endemic. Taylor (unpublished notes) finds that this 

Dalla Torre name applies to a sibling relative of M. castanea which has winged queens, 

also found on North I. 
Brachyponera luteipes (Mayr) [=B. chinensis (Emery) ?]. Introduced from Indo-Australian 

area. Brown (1958), Taylor (1961). 
Hypoponera eduardi (Forel) [=Ponera antipodum Forel]. Introduced, ultimate origin probably 

Africa. Brown (1958), Taylor (1967). 

Ponera leae Forel. Introduced from E. Australia or Tasmania. Taylor (1960, 1966). 
Phddole megacephala (Fabricius). Introduced, ultimate origin Africa. Taylor (1961). 
Pheidole variabilis Mayr. Introduced from E. Australia. Taylor (1961). 

Pheidole vigilans Fr. Smith (=yarrenris Forel). Introduced from E. Australia. Cumber (1959). 
Huberia striata (Fr. Smith). Endemic. Brown (1958). [Note: the genus Huberia is endemic 

to New Zealand.] 

Huberia brouni Forel. Endemic. Brown (1958). 

Tetramorium grassii Emery. Introduced from South Africa. Brown (1958). 
Tetramorium guineense (Fabricius). Introduced, ultimate origin Africa. Taylor (1961). 
Chelaner antarcticus (White). Endemic. Brown (1958). 
Chelaner smithi Forel. Endemic. Brown (1958), Taylor (1959). 
Monomorium orientals Mayr. Introduced from Old World tropics. Brown (1958). 
Monomorium pharaonis (L.). Introduced, ultimate origin Old World tropics. Brown (1958), 

Taylor (1961). 

Mayriella abstinens Forel. Introduced from E. Australia. Taylor (1961). 
Orectognathus antennatus Fr Smith. Probably introduced from Australia. Brown (1958). 
Strumigenys perplexa (Fr. Smith). Probably introduced from E. Australia. Brown (1958). 
Iridomyrmex darwinianus (Forel). Introduced from E. Australia. Taylor (1959). 
Iridomyrmex glaber (Mayr). Probably introduced from E. Australia. Brown (1958). 
Technomyrmex albipes (Fr. Smith). Introduced, ultimate origin Asia. Brown (1958). 
Prolasius advena (Fr. Smith). Endemic. Brown (1958). 

Camponotus (Colobopsis') newzealandicus Donisthorpe. Endemic? Brown (1958). 
Paratrechina vaga Forel. Introduced, ultimate origin Indo-Australian area. Brown (1958). 


Blackburn, T. & W. F. Kirby. 1880. Notes on species of aculeate Hymenoptera occurring 

in the Hawaiian Islands. Ent. Man. Mag. 17 : 89. 
Brown, W. L. 1949a. Revision of the ant tribe Dacetini: I. Fauna of Japan, China and 

Taiwan. Mushi 20 : 1-25. 
1949b. Revision of the ant tribe Dacetini : III. Epitritus Emery and Quadristruma 

new genus. Trans. Amer. Ent. Sac. 75: 43-51. 
1954. The ant genus Strumigenys Fred. Smith in the Ethiopian and Malagasy Regions. 

Bull. Mus. Comp. Zool. Harv. 112: 3-34. 

1958. A review of the ants of New Zealand. Acta Hymenopt. 1 : 1-50. 
1960. Contributions toward a reclassification of the Formicidae. HI. Tribe Ambly- 

oponini (Hymenoptera). Bull. Mus. Comp. Zool. Harv. 122: 145-230. 
1964. The ant genus Smithistruma: A first supplement to the world revision (Hyme- 
noptera : Formicidae). Trans. Am. Ent. Soc. 89: 183-200. 

Brown, W. L. & W. W. Kempf. 1960. A world revision of the ant tribe Basicerotini. 
Studia Ent. (n. series) 3: 161-250. 

Pac. Ins. Mon. 

Brown, W. L. & E. O. Wilson. 1959. The evolution of the dacetine ants. Quart. Rev. 
Biol. 34: 278-94. 

Bryan, E. H. 1935. Insects from Rabbit Island. Proc. Hawn. Ent. Sac. 9: 39-44. 

Butler, G. D. 1961. Insects and other arthropods from Laysan Island. Ibid. 17: 379-87. 

Carter, W. 1936. (Notes). Ibid. 9: 143. 

1957. (Exhibitions). Proc. Hawn. Ent. Sac. for 1956, p. 183. 

Cheesman, L. E. & W. C. Crawley. 1928. A contribution towards the insect fauna of 
French Oceania, pt. 3, Formicidae. Ann. Mag. Nat. Hist. Ser. 10, 2: 514-25. 

Chilson, L. E. 1953. Insect records from Johnston Island. Proc. Hawn. Ent. Soc. 15: 81- 

Clagg, C. F. 1956. (Exhibitions). Proc. Hawn. Ent. Soc. for 1955, p. 4. 
1957. (Exhibitions). Proc. Hawn. Ent. Soc. for 1956, p. 197. 

Creighton, W. S. 1930. The New World species of the genus Solenopsis (Hymenop. For- 
micidae). Proc. Amer. Acad. Arts Sci. 66: 39-151. 

Cumber, R. A. 1959. Distributional and biological notes on sixteen North Island species 
of Formicidae (Hymenoptera). New Zeal. Ent. 2 (4): 10-14. 

Donisthorpe, H. 1932. On the identity of Smith's types of Formicidae (Hymenoptera) 
collected by Alfred Russell Wallace in the Malay Archipelago, with the descriptions 
of two new species. Ann. Mag. Nat. Hist. Ser. 10, 10 : 441-76. 

Emery, C. 1899. Ergebnesse einer Reise nach dem Pacific (Schauinsland 1896-97). Zool. 
Jahrb., Abt. Syst. 12: 438-40. 

1900. Formicidarum species novae vel minus cognitae...Nova Guinea. Termeszetr, Fiiz. 
23: 310-38. 

1914a. Formiche d'Australia e di Samoa raccolte dal Prof. Silvestri nel 1913. Boll. 

Lab. Zool. Gen. Agrar. Port id 8: 179-86. 
1914b. Les fourmis de la Nouvelle-Caledonie et des iles Loyalty. In Sarasin and Roux, 

Nova Caledonia, Zool. 1 : 393-435. 

1921. Hymenoptera : Formicidae : Myrmicinae. Gen. Ins. 174 : 1-397. 
1925. Hymenoptera : Formicidae : Formicinae. Ibid. 183 : 1-302. 
Ettershank, G. 1965. A generic revision of the world Myrmicinae related to Solenopsis 

and Pheidologeton (Hymenoptera: Formicidae). Aust. J. Zool. 14: 73-171. 
Ford, E. J. 1959. (Exhibitions and Notes). Proc. Hawn. Ent. Soc. for 1958, p. 18. 
Forel, A. 1899. Formicidae. Fauna Hawaiiensis, Cambridge, pp. 116-122. 

1901. Formiciden aus dem Bismark-Archipel. Mitt. Zool. Mas. Berlin 2 : 5-37. 
1914. Le genre Camponotus et les genres voisins. Rev. Suisse Zool. 22 : 257-75. 

Illingworth, J. F. 1917. Economic aspects of our predaceous ant (Pheidole megacephala} . 
Proc. Hawn. Ent. Soc. 4: 349-68. 

Ito, K. 1943. Seasonal trend in the development of the sex forms of the fire ant, Sole- 
nopsis geminata (Fabr.) var. rufa (Jerdon), in the Pineapple fields of Oahu. Proc. 
Hawn. Ent. Soc. 11 : 171-75. 

Kempf, W. W. 1962. Miscellaneous studies on Neotropical ants (Hym. Formicidae). Stu- 
dia Ent. 5: 1-38. 

Mann, W. M. 1921. The ants of the Fiji Islands. Bull. Mus. Comp. Zool. Harv. 64: 399-499. 

Mayr, G. 1866. Myrmecologische Beitrage. Sitz. Akad. Wiss. Wien 53: 509. 
1870. Neue Formiciden. Verh. Zool.-bot. Ges. Wien 20: 939-96. 
1876. Die australischen Ameisen. /. Mus. Godeffroy Hamburg 12: 56-115. 

Wilson & Taylor : Ants of Polynesia 

Pemberton, C. E. 1944. (Notes and exhibitions). Proc. Hawn. Ent. Soc. 12: 25. 

Phillips, J. S. 1934. Fire ants in dry areas. Ibid. 8: 515. 

Santschi, F. 1919. Cinq notes myrmecologiques. Bull Vaud. Sci. Nat. 52: 325-50. 

1928. Formicidae. Insects of Samoa 5 (1): 41-58. 
Smith, M. R. 1944. Ants of the genus Cardiocondyla Emery in the United States. Proc. 

Wash. Ent. Soc. 46: 30-41. 

1947. A generic and subgeneric synopsis of the United States ants based on the work- 
ers (Hymenoptera: Formicidae). Amer. Midi. Nat. 37: 521-647. 
1951. Formicidae, In Synoptic Catalogue of Hymenoptera north of Mexico. U. S. Dept. 

Agr. Man. 2: 839. 

1957. A contribution to the taxonomy, distribution and biology of the vagrant ant, 

Plagiolepis alluaudi Emery (Hymenoptera, Formicidae). J. N.Y. Ent. Soc. 65: 159-98. 

Suehiro, A 1960. Insects and other arthropods from Midway Atoll. Proc. Hawn. Ent. Soc. 

27 (2): 289-98. 
Swezey, O. H. 1943. Notes and exhibitions. Ibid. 12 : 25. 

1945. Insects associated with orchids. Ibid. 12: 343-403. 

Taylor, R. W. 1959. The Australian ant Iridomyrmex darwinianus (Forel) recorded from 
New Zealand. New Zeal. Ent. 2 (4): 18-19. 

1960. Taxonomic notes on the ants Ponera leae Forel and Ponera norfolkensis (Wheel- 
er) (Hymenoptera-Formicidae). Pacific Sci. 14: 178-80. 

1961. Notes and new records of exotic ants introduced into New Zealand. New Zeal. 
Ent. 2 (6): 1-10. 

1967. A monographic revision of the ant genus Ponera Latreille (Hymenoptera-Formi- 
cidae). Pac. Ins. Man. 13: 1-112. 
Taylor, R. W. & E O Wilson. 1961. Ants from three remote oceanic islands. Psyche 68 

(4): 137-44. 
Timberlake, P. H. 1926. Hymenoptera. In "Insects of Hawaii, Johnston Island and Wake 

Islands," ed. E. H. Bryan. Bull. Bishop Mus. 31 : 17-19. 
Wheeler, W. M. 1908. Ants from Moorea, Society Islands. Bull. Amer. Mus. Nat. Hist. 

24: 165-67. 
1909. Ants collected by Prof F. Silvestri in the Hawaiian Islands. Boll. Lab. Zool. 

Gen. Agrar. Portici 3 : 269-72. 
1922. Formicidae from Easter Island and Juan Fernandez. In Skottsberg, ed., Nat. 

Hist. Juan Fernandez and Easter Island 2 : 317-19. 

1932a. Ants from the Society Islands. Bull. B. P. Bishop Mus. 113: 13-19. 
1932b. Ants of the Marquesas Islands. Ibid. 98: 155-63. 
1933a. Formicidae of the Templeton Crocker Expedition. Proc. Calif. Acad. Sci. Ser. 

4, 21 : 57-64. 

1933b. Three obscure genera of ponerine ants. Amer. Mus. Nat. Hist. Novit. 672: 1-23. 
1933c. An ant new to the fauna of the Hawaiian Islands. Proc. Hawn. Ent. Soc. 8 : 

1933d. A new species of Ponera and other records of ants from the Marquesas Islands. 

Bull. B. P. Bishop Mus. 114 : 141-44. 

1934a. Revised list of Hawaiian ants. B. P. Bishop Mus., Occ. Pap. 10: 3-21. 
1934b. Check list of the ants of Oceania. Ibid. 11 : 3-56. 
1936. Ants from the Society, Austral, Tuamotu and Mangareva Islands. Ibid. 12: 3- 

Pac. Ins. Mon. 

Williams, F. X. 1937. (Notes and exhibitions). Proc. Hawn. Ent. Soc. 9: 375. 
Wilson, E. O. 1953. The ecology of some North American dacetine ants. Ann. Ent. Soc. 
Amer. 46: 479-95. 

1957. The tennis and selenophora groups of the ant genus Ponera. Bull. Mus. Comp. Zool. 
Harv. 116: 355-86. 

1958. Studies on the ant fauna of Melanesia I-IV. Ibid. 118: 101-53; 119: 303-71. 

1959. Studies on the ant fauna of Melanesia V. Ibid. 120: 483-510. 

1960. The ants of Rennell and Bellona Islands, British Solomon Islands. Nat. Hist, 
of Rennell I., Brit Solomon Is., (Copenhagen) 4: 13-23. 

Wilson, E. O. & R. W. Taylor. 1967. An estimate of the potential evolutionary increase 

of species density in the Polynesian ant fauna. Evolution 11 : 1-10. 
Zimmerman, E. C. 1940. Argentine ant in Hawaii. Proc. Hawn. Ent. Soc. 11 : 108. 

1953. Notes and exhibitions. Ibid. 15: 10. 
Van Zwaluwenburg, R. H. 1943. The insects of Canton Island. Ibid. 11 : 300-12. 


New names are indicated by bold face type, synonyms by italics. 

















atomus (Oligomyrmex). 
atomus (Paratrechina)... 





























16, 42 crassipilis 90 

52 Cryptopone 

77 cubensis 75 

21 Dacetini 13 

57 dahlii 60 

18, 82, 102 decipiens 

18, 85 dentatus < 

19 denticulata 58 

42 destructor 15, 64, 66 

....7, 18, 78 detorquens 84 

28 Dolichoderinae 10, 78 

28 dubia 13, 36 

32 dyak 38 

84 Ectomomyrmex 21 

15, 62 emeryi 16, 53 

92 emmae 13, 42 

77 Eurhopalothrix 34 

...14, 60, 62 exigua 18, 86 

89 exsulans 7, 16, 74 

16, 43 falcata 30 

85 falcigera 30 

80 fervens 7, 8, 17, 45 

75 flavolimbatus 17, 96, 98 

88 floricola 15, 64, 102 

48 foreli 85 

, 18, 87, 102 Formicinae 10,84 

92 fossulatum 15, 65 

58 fullawayi 82 

75 fuscipennis 31 

96 geminata 14, 58, 102 

97 germaini 74 

70 gleadowi 28 

93 godeffroyi 14, 36 

75 gracilis 78 

8, 53 gracillimum 66, 102 

45 graeffei 10, 32 

10, 33 guineense 15, 71, 73, 102 

99 guppyi 96 

62 haematodes 31 

93 haematodus 31 

17, 93 hawaiiensis (Cardiocondyla) 56 

60 hawaiiensis (Paratrechina) 88 

">4 hawaiiensis (Camponotus) 95 

.6 heeri 92 

12, 26 hirsutus 77 

17, 96 humilis 18, 79 

97 Hypoponera 26 

Pac. Ins. Mon. 






Iridomyrmex .... 
























melanocephalum . 




minutum (Monomorium). 

minutum (Tapinoma) 


mumfordi (Hypoponera)... 
mumfordi (Smithistruma) 

my ops 




nautarum (Camponotus) ... 
nautarum (Hypoponera) 




12, 23, 25 




2, 11, 21 








11, 19 

15, 66 



14,38, 41 


12, 24 

18,87, 102 

18, 84 



5, 85 


14, 38 




. 8, 16, 46, 59, 80, 102 

18, 80, 102 

13, 35 

11, 20 



18, 89 

15, 67 

18, 82 

15, 64 


5, 13, 36 



10, 13, 34 




17, 94 


17, 97 











oceanica (Pheidole) 

oceanicus (Anochetus) 







pacifica (Platythyrea) 

pacifica (Vollenhovia) 

pacificum (Tetramorium) 

























rapaense (Chelaner ; Monomorium) 









8, 17, 48, 53 






.. 14, 56, 58 
... 7, 15, 72 


14, 60 


8, 87 

11, 20 




15, 68, 102 


....8, 13, 42 

17, 85 


99, 102 

19, 23 

10, 19 


13, 34 




rotumana (Polyrhachis) . . 
rotumanus (Camponotus) 

rufifrons 96 

samoanum (Monomorium) 67 

samoanus (Adelomyrmex) 14, 77 

samoensis (Camponotus) 93 

samoensis (Vollenhovia) 2, 14, 57 

sanctaecrucis 93 

scabra 74 

schauinslandi 28 

sechellense 65 

seini 33 

sexspinosa 16, 52 

sharp! 89 

silvestrii 5, 10, 33 

simillimum (Tetramorium) 15, 73, 102 

simillimus (Ectomomyrmex) 21 

simillimus (Odontomachus) 7, 10, 31 

sinensis 34 

skottsbergi 88 

Smithistruma 13, 36 

Solenopsis 58 

speculare 64 

stigma 11, 22 

stigmatica (Paratrechina) 90 

stigmatica (Rogeria) 75, 76 

striatidens 14, 70 

Strumigenys 13, 36 

subcoenum 66 

sublevinodis 16, 75, 76 

swezeyi 5, 12, 19,25 

syscena 24 

Syscia 33 

szalayi 13, 41 

tahitiana 45 

tahitiensis 5, 14, 62 






















vitiensis (Paratrechina) 

vitiensis (Solenopsis) 

vitiensis (Technomyrmex) 




wheeleri (Epitritus) 

wheeleri (Quadristruma) 

wheeleri (Zatapinoma) 







zwaluwenburgi ( Amblyopone) . 
zwaluwenburgi (Hypoponera).. 

15, 69 




.12, 24, 25 

11, 23 



15, 74 



33, 34 

16, 50 


18, 88, 90 


17, 95 










12, 24, 26 

16, 56 




..5, 11, 19 
. 5, 12, 29